1 \input texinfo @c -*-texinfo-*-
2 @c Copyright 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998
3 @c Free Software Foundation, Inc.
6 @c makeinfo ignores cmds prev to setfilename, so its arg cannot make use
7 @c of @set vars. However, you can override filename with makeinfo -o.
13 @settitle Debugging with @value{GDBN}
16 @settitle Debugging with @value{GDBN} (@value{TARGET})
19 @setchapternewpage odd
30 @c readline appendices use @vindex
33 @c !!set GDB manual's edition---not the same as GDB version!
36 @c !!set GDB manual's revision date
37 @set DATE January 1994
39 @c THIS MANUAL REQUIRES TEXINFO-2 macros and info-makers to format properly.
42 @c This is a dir.info fragment to support semi-automated addition of
43 @c manuals to an info tree. zoo@cygnus.com is developing this facility.
46 * Gdb: (gdb). The @sc{gnu} debugger.
53 This file documents the @sc{gnu} debugger @value{GDBN}.
56 This is Edition @value{EDITION}, @value{DATE},
57 of @cite{Debugging with @value{GDBN}: the @sc{gnu} Source-Level Debugger}
58 for @value{GDBN} Version @value{GDBVN}.
60 Copyright (C) 1988, 1989, 1990, 1991, 1992, 1993, 1994, 1995
61 Free Software Foundation, Inc.
63 Permission is granted to make and distribute verbatim copies of
64 this manual provided the copyright notice and this permission notice
65 are preserved on all copies.
68 Permission is granted to process this file through TeX and print the
69 results, provided the printed document carries copying permission
70 notice identical to this one except for the removal of this paragraph
71 (this paragraph not being relevant to the printed manual).
74 Permission is granted to copy and distribute modified versions of this
75 manual under the conditions for verbatim copying, provided also that the
76 entire resulting derived work is distributed under the terms of a
77 permission notice identical to this one.
79 Permission is granted to copy and distribute translations of this manual
80 into another language, under the above conditions for modified versions.
84 @title Debugging with @value{GDBN}
85 @subtitle The @sc{gnu} Source-Level Debugger
87 @subtitle (@value{TARGET})
90 @subtitle Edition @value{EDITION}, for @value{GDBN} version @value{GDBVN}
91 @subtitle @value{DATE}
92 @author Richard M. Stallman and Roland H. Pesch
96 \hfill (Send bugs and comments on @value{GDBN} to bug-gdb\@prep.ai.mit.edu.)\par
97 \hfill {\it Debugging with @value{GDBN}}\par
98 \hfill \TeX{}info \texinfoversion\par
99 \hfill doc\@cygnus.com\par
103 @vskip 0pt plus 1filll
104 Copyright @copyright{} 1988, 1989, 1990, 1991, 1992, 1993, 1994, 1995
105 Free Software Foundation, Inc.
107 Published by the Free Software Foundation @*
108 59 Temple Place - Suite 330, @*
109 Boston, MA 02111-1307 USA @*
110 Printed copies are available for $20 each. @*
111 ISBN 1-882114-11-6 @*
113 Permission is granted to make and distribute verbatim copies of
114 this manual provided the copyright notice and this permission notice
115 are preserved on all copies.
117 Permission is granted to copy and distribute modified versions of this
118 manual under the conditions for verbatim copying, provided also that the
119 entire resulting derived work is distributed under the terms of a
120 permission notice identical to this one.
122 Permission is granted to copy and distribute translations of this manual
123 into another language, under the above conditions for modified versions.
129 @top Debugging with @value{GDBN}
131 This file describes @value{GDBN}, the @sc{gnu} symbolic debugger.
133 This is Edition @value{EDITION}, @value{DATE}, for @value{GDBN} Version
137 * Summary:: Summary of @value{GDBN}
139 * Sample Session:: A sample @value{GDBN} session
142 * Invocation:: Getting in and out of @value{GDBN}
143 * Commands:: @value{GDBN} commands
144 * Running:: Running programs under @value{GDBN}
145 * Stopping:: Stopping and continuing
146 * Stack:: Examining the stack
147 * Source:: Examining source files
148 * Data:: Examining data
150 * Languages:: Using @value{GDBN} with different languages
153 * C:: C language support
155 @c remnant makeinfo bug, blank line needed after two end-ifs?
157 * Symbols:: Examining the symbol table
158 * Altering:: Altering execution
159 * GDB Files:: @value{GDBN} files
160 * Targets:: Specifying a debugging target
161 * Controlling GDB:: Controlling @value{GDBN}
162 * Sequences:: Canned sequences of commands
164 * Emacs:: Using @value{GDBN} under @sc{gnu} Emacs
167 * GDB Bugs:: Reporting bugs in @value{GDBN}
168 * Command Line Editing:: Facilities of the readline library
169 * Using History Interactively::
171 @c * Renamed Commands::
173 @ifclear PRECONFIGURED
174 * Formatting Documentation:: How to format and print @value{GDBN} documentation
175 * Installing GDB:: Installing GDB
183 @unnumbered Summary of @value{GDBN}
185 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
186 going on ``inside'' another program while it executes---or what another
187 program was doing at the moment it crashed.
189 @value{GDBN} can do four main kinds of things (plus other things in support of
190 these) to help you catch bugs in the act:
194 Start your program, specifying anything that might affect its behavior.
197 Make your program stop on specified conditions.
200 Examine what has happened, when your program has stopped.
203 Change things in your program, so you can experiment with correcting the
204 effects of one bug and go on to learn about another.
208 You can use @value{GDBN} to debug programs written in C or C++.
209 @c "MOD2" used as a "miscellaneous languages" flag here.
210 @c This is acceptable while there is no real doc for Chill and Pascal.
212 For more information, see @ref{Support,,Supported languages}.
215 For more information, see @ref{C,,C and C++}.
217 Support for Modula-2 and Chill is partial. For information on Modula-2,
218 see @ref{Modula-2,,Modula-2}. There is no further documentation on Chill yet.
220 Debugging Pascal programs which use sets, subranges, file variables, or nested
221 functions does not currently work. @value{GDBN} does not support
222 entering expressions, printing values, or similar features using Pascal syntax.
227 @value{GDBN} can be used to debug programs written in Fortran, although
228 it does not yet support entering expressions, printing values, or
229 similar features using Fortran syntax. It may be necessary to refer to
230 some variables with a trailing underscore.
235 * Free Software:: Freely redistributable software
236 * Contributors:: Contributors to GDB
240 @unnumberedsec Free software
242 @value{GDBN} is @dfn{free software}, protected by the @sc{gnu}
243 General Public License
244 (GPL). The GPL gives you the freedom to copy or adapt a licensed
245 program---but every person getting a copy also gets with it the
246 freedom to modify that copy (which means that they must get access to
247 the source code), and the freedom to distribute further copies.
248 Typical software companies use copyrights to limit your freedoms; the
249 Free Software Foundation uses the GPL to preserve these freedoms.
251 Fundamentally, the General Public License is a license which says that
252 you have these freedoms and that you cannot take these freedoms away
256 @unnumberedsec Contributors to GDB
258 Richard Stallman was the original author of GDB, and of many other @sc{gnu}
259 programs. Many others have contributed to its development. This
260 section attempts to credit major contributors. One of the virtues of
261 free software is that everyone is free to contribute to it; with
262 regret, we cannot actually acknowledge everyone here. The file
263 @file{ChangeLog} in the @value{GDBN} distribution approximates a blow-by-blow
266 Changes much prior to version 2.0 are lost in the mists of time.
269 @emph{Plea:} Additions to this section are particularly welcome. If you
270 or your friends (or enemies, to be evenhanded) have been unfairly
271 omitted from this list, we would like to add your names!
274 So that they may not regard their long labor as thankless, we
275 particularly thank those who shepherded GDB through major releases:
276 Stan Shebs (release 4.14),
277 Fred Fish (releases 4.13, 4.12, 4.11, 4.10, and 4.9),
278 Stu Grossman and John Gilmore (releases 4.8, 4.7, 4.6, 4.5, and 4.4),
279 John Gilmore (releases 4.3, 4.2, 4.1, 4.0, and 3.9);
280 Jim Kingdon (releases 3.5, 3.4, and 3.3);
281 and Randy Smith (releases 3.2, 3.1, and 3.0).
282 As major maintainer of @value{GDBN} for some period, each
283 contributed significantly to the structure, stability, and capabilities
284 of the entire debugger.
286 Richard Stallman, assisted at various times by Peter TerMaat, Chris
287 Hanson, and Richard Mlynarik, handled releases through 2.8.
290 Michael Tiemann is the author of most of the @sc{gnu} C++ support in GDB,
291 with significant additional contributions from Per Bothner. James
292 Clark wrote the @sc{gnu} C++ demangler. Early work on C++ was by Peter
293 TerMaat (who also did much general update work leading to release 3.0).
296 @value{GDBN} 4 uses the BFD subroutine library to examine multiple
297 object-file formats; BFD was a joint project of David V.
298 Henkel-Wallace, Rich Pixley, Steve Chamberlain, and John Gilmore.
300 David Johnson wrote the original COFF support; Pace Willison did
301 the original support for encapsulated COFF.
303 Brent Benson of Harris Computer Systems contributed DWARF 2 support.
305 Adam de Boor and Bradley Davis contributed the ISI Optimum V support.
306 Per Bothner, Noboyuki Hikichi, and Alessandro Forin contributed MIPS
308 Jean-Daniel Fekete contributed Sun 386i support.
309 Chris Hanson improved the HP9000 support.
310 Noboyuki Hikichi and Tomoyuki Hasei contributed Sony/News OS 3 support.
311 David Johnson contributed Encore Umax support.
312 Jyrki Kuoppala contributed Altos 3068 support.
313 Jeff Law contributed HP PA and SOM support.
314 Keith Packard contributed NS32K support.
315 Doug Rabson contributed Acorn Risc Machine support.
316 Bob Rusk contributed Harris Nighthawk CX-UX support.
317 Chris Smith contributed Convex support (and Fortran debugging).
318 Jonathan Stone contributed Pyramid support.
319 Michael Tiemann contributed SPARC support.
320 Tim Tucker contributed support for the Gould NP1 and Gould Powernode.
321 Pace Willison contributed Intel 386 support.
322 Jay Vosburgh contributed Symmetry support.
324 Rich Schaefer and Peter Schauer helped with support of SunOS shared
327 Jay Fenlason and Roland McGrath ensured that @value{GDBN} and GAS agree about
328 several machine instruction sets.
330 Patrick Duval, Ted Goldstein, Vikram Koka and Glenn Engel helped develop
331 remote debugging. Intel Corporation, Wind River Systems, AMD, and ARM
332 contributed remote debugging modules for the i960, VxWorks, A29K UDI,
333 and RDI targets, respectively.
335 Brian Fox is the author of the readline libraries providing
336 command-line editing and command history.
338 Andrew Beers of SUNY Buffalo wrote the language-switching code,
340 the Modula-2 support,
342 and contributed the Languages chapter of this manual.
344 Fred Fish wrote most of the support for Unix System Vr4.
346 He also enhanced the command-completion support to cover C++ overloaded
350 Hitachi America, Ltd. sponsored the support for Hitachi microprocessors.
352 Kung Hsu, Jeff Law, and Rick Sladkey added support for hardware
355 Michael Snyder added support for tracepoints.
357 Stu Grossman wrote gdbserver.
359 Jim Kingdon, Peter Schauer, Ian Taylor, and Stu Grossman made
360 nearly innumerable bug fixes and cleanups throughout GDB.
362 Cygnus Solutions has sponsored GDB maintenance and much of its
363 development since 1991.
367 @chapter A Sample @value{GDBN} Session
369 You can use this manual at your leisure to read all about @value{GDBN}.
370 However, a handful of commands are enough to get started using the
371 debugger. This chapter illustrates those commands.
374 In this sample session, we emphasize user input like this: @b{input},
375 to make it easier to pick out from the surrounding output.
378 @c FIXME: this example may not be appropriate for some configs, where
379 @c FIXME...primary interest is in remote use.
381 One of the preliminary versions of @sc{gnu} @code{m4} (a generic macro
382 processor) exhibits the following bug: sometimes, when we change its
383 quote strings from the default, the commands used to capture one macro
384 definition within another stop working. In the following short @code{m4}
385 session, we define a macro @code{foo} which expands to @code{0000}; we
386 then use the @code{m4} built-in @code{defn} to define @code{bar} as the
387 same thing. However, when we change the open quote string to
388 @code{<QUOTE>} and the close quote string to @code{<UNQUOTE>}, the same
389 procedure fails to define a new synonym @code{baz}:
398 @b{define(bar,defn(`foo'))}
402 @b{changequote(<QUOTE>,<UNQUOTE>)}
404 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
407 m4: End of input: 0: fatal error: EOF in string
411 Let us use @value{GDBN} to try to see what is going on.
414 $ @b{@value{GDBP} m4}
415 @c FIXME: this falsifies the exact text played out, to permit smallbook
416 @c FIXME... format to come out better.
417 @value{GDBN} is free software and you are welcome to distribute copies
418 of it under certain conditions; type "show copying" to see
420 There is absolutely no warranty for @value{GDBN}; type "show warranty"
423 @value{GDBN} @value{GDBVN}, Copyright 1995 Free Software Foundation, Inc...
428 @value{GDBN} reads only enough symbol data to know where to find the
429 rest when needed; as a result, the first prompt comes up very quickly.
430 We now tell @value{GDBN} to use a narrower display width than usual, so
431 that examples fit in this manual.
434 (@value{GDBP}) @b{set width 70}
438 We need to see how the @code{m4} built-in @code{changequote} works.
439 Having looked at the source, we know the relevant subroutine is
440 @code{m4_changequote}, so we set a breakpoint there with the @value{GDBN}
441 @code{break} command.
444 (@value{GDBP}) @b{break m4_changequote}
445 Breakpoint 1 at 0x62f4: file builtin.c, line 879.
449 Using the @code{run} command, we start @code{m4} running under @value{GDBN}
450 control; as long as control does not reach the @code{m4_changequote}
451 subroutine, the program runs as usual:
454 (@value{GDBP}) @b{run}
455 Starting program: /work/Editorial/gdb/gnu/m4/m4
463 To trigger the breakpoint, we call @code{changequote}. @value{GDBN}
464 suspends execution of @code{m4}, displaying information about the
465 context where it stops.
468 @b{changequote(<QUOTE>,<UNQUOTE>)}
470 Breakpoint 1, m4_changequote (argc=3, argv=0x33c70)
472 879 if (bad_argc(TOKEN_DATA_TEXT(argv[0]),argc,1,3))
476 Now we use the command @code{n} (@code{next}) to advance execution to
477 the next line of the current function.
481 882 set_quotes((argc >= 2) ? TOKEN_DATA_TEXT(argv[1])\
486 @code{set_quotes} looks like a promising subroutine. We can go into it
487 by using the command @code{s} (@code{step}) instead of @code{next}.
488 @code{step} goes to the next line to be executed in @emph{any}
489 subroutine, so it steps into @code{set_quotes}.
493 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
495 530 if (lquote != def_lquote)
499 The display that shows the subroutine where @code{m4} is now
500 suspended (and its arguments) is called a stack frame display. It
501 shows a summary of the stack. We can use the @code{backtrace}
502 command (which can also be spelled @code{bt}), to see where we are
503 in the stack as a whole: the @code{backtrace} command displays a
504 stack frame for each active subroutine.
507 (@value{GDBP}) @b{bt}
508 #0 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
510 #1 0x6344 in m4_changequote (argc=3, argv=0x33c70)
512 #2 0x8174 in expand_macro (sym=0x33320) at macro.c:242
513 #3 0x7a88 in expand_token (obs=0x0, t=209696, td=0xf7fffa30)
515 #4 0x79dc in expand_input () at macro.c:40
516 #5 0x2930 in main (argc=0, argv=0xf7fffb20) at m4.c:195
520 We step through a few more lines to see what happens. The first two
521 times, we can use @samp{s}; the next two times we use @code{n} to avoid
522 falling into the @code{xstrdup} subroutine.
526 0x3b5c 532 if (rquote != def_rquote)
528 0x3b80 535 lquote = (lq == nil || *lq == '\0') ? \
529 def_lquote : xstrdup(lq);
531 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
534 538 len_lquote = strlen(rquote);
538 The last line displayed looks a little odd; we can examine the variables
539 @code{lquote} and @code{rquote} to see if they are in fact the new left
540 and right quotes we specified. We use the command @code{p}
541 (@code{print}) to see their values.
544 (@value{GDBP}) @b{p lquote}
545 $1 = 0x35d40 "<QUOTE>"
546 (@value{GDBP}) @b{p rquote}
547 $2 = 0x35d50 "<UNQUOTE>"
551 @code{lquote} and @code{rquote} are indeed the new left and right quotes.
552 To look at some context, we can display ten lines of source
553 surrounding the current line with the @code{l} (@code{list}) command.
559 535 lquote = (lq == nil || *lq == '\0') ? def_lquote\
561 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
564 538 len_lquote = strlen(rquote);
565 539 len_rquote = strlen(lquote);
572 Let us step past the two lines that set @code{len_lquote} and
573 @code{len_rquote}, and then examine the values of those variables.
577 539 len_rquote = strlen(lquote);
580 (@value{GDBP}) @b{p len_lquote}
582 (@value{GDBP}) @b{p len_rquote}
587 That certainly looks wrong, assuming @code{len_lquote} and
588 @code{len_rquote} are meant to be the lengths of @code{lquote} and
589 @code{rquote} respectively. We can set them to better values using
590 the @code{p} command, since it can print the value of
591 any expression---and that expression can include subroutine calls and
595 (@value{GDBP}) @b{p len_lquote=strlen(lquote)}
597 (@value{GDBP}) @b{p len_rquote=strlen(rquote)}
602 Is that enough to fix the problem of using the new quotes with the
603 @code{m4} built-in @code{defn}? We can allow @code{m4} to continue
604 executing with the @code{c} (@code{continue}) command, and then try the
605 example that caused trouble initially:
611 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
618 Success! The new quotes now work just as well as the default ones. The
619 problem seems to have been just the two typos defining the wrong
620 lengths. We allow @code{m4} exit by giving it an EOF as input:
624 Program exited normally.
628 The message @samp{Program exited normally.} is from @value{GDBN}; it
629 indicates @code{m4} has finished executing. We can end our @value{GDBN}
630 session with the @value{GDBN} @code{quit} command.
633 (@value{GDBP}) @b{quit}
638 @chapter Getting In and Out of @value{GDBN}
640 This chapter discusses how to start @value{GDBN}, and how to get out of it.
644 type @samp{@value{GDBP}} to start GDB.
646 type @kbd{quit} or @kbd{C-d} to exit.
650 * Invoking GDB:: How to start @value{GDBN}
651 * Quitting GDB:: How to quit @value{GDBN}
652 * Shell Commands:: How to use shell commands inside @value{GDBN}
656 @section Invoking @value{GDBN}
659 For details on starting up @value{GDBP} as a
660 remote debugger attached to a Hitachi microprocessor, see @ref{Hitachi
661 Remote,,@value{GDBN} and Hitachi Microprocessors}.
664 Invoke @value{GDBN} by running the program @code{@value{GDBP}}. Once started,
665 @value{GDBN} reads commands from the terminal until you tell it to exit.
667 You can also run @code{@value{GDBP}} with a variety of arguments and options,
668 to specify more of your debugging environment at the outset.
671 The command-line options described here are designed
672 to cover a variety of situations; in some environments, some of these
673 options may effectively be unavailable.
676 The most usual way to start @value{GDBN} is with one argument,
677 specifying an executable program:
680 @value{GDBP} @var{program}
685 You can also start with both an executable program and a core file
689 @value{GDBP} @var{program} @var{core}
692 You can, instead, specify a process ID as a second argument, if you want
693 to debug a running process:
696 @value{GDBP} @var{program} 1234
700 would attach @value{GDBN} to process @code{1234} (unless you also have a file
701 named @file{1234}; @value{GDBN} does check for a core file first).
703 Taking advantage of the second command-line argument requires a fairly
704 complete operating system; when you use @value{GDBN} as a remote debugger
705 attached to a bare board, there may not be any notion of ``process'',
706 and there is often no way to get a core dump.
709 You can run @code{gdb} without printing the front material, which describes
710 @value{GDBN}'s non-warranty, by specifying @code{-silent}:
713 @value{GDBP} @var{-silent}
717 You can further control how @value{GDBN} starts up by using command-line
718 options. @value{GDBN} itself can remind you of the options available.
728 to display all available options and briefly describe their use
729 (@samp{@value{GDBP} -h} is a shorter equivalent).
731 All options and command line arguments you give are processed
732 in sequential order. The order makes a difference when the
733 @samp{-x} option is used.
739 * Remote Serial:: @value{GDBN} remote serial protocol
742 * i960-Nindy Remote:: @value{GDBN} with a remote i960 (Nindy)
745 * UDI29K Remote:: The UDI protocol for AMD29K
746 * EB29K Remote:: The EBMON protocol for AMD29K
749 * VxWorks Remote:: @value{GDBN} and VxWorks
752 * ST2000 Remote:: @value{GDBN} with a Tandem ST2000
755 * Hitachi Remote:: @value{GDBN} and Hitachi Microprocessors
758 * MIPS Remote:: @value{GDBN} and MIPS boards
761 * Sparclet Remote:: @value{GDBN} and Sparclet boards
764 * Simulator:: Simulated CPU target
767 @c remnant makeinfo bug requires this blank line after *two* end-ifblahs:
769 * File Options:: Choosing files
770 * Mode Options:: Choosing modes
778 @subsection Choosing files
781 When @value{GDBN} starts, it reads any arguments other than options as
782 specifying an executable file and core file (or process ID). This is
783 the same as if the arguments were specified by the @samp{-se} and
784 @samp{-c} options respectively. (@value{GDBN} reads the first argument
785 that does not have an associated option flag as equivalent to the
786 @samp{-se} option followed by that argument; and the second argument
787 that does not have an associated option flag, if any, as equivalent to
788 the @samp{-c} option followed by that argument.)
791 When @value{GDBN} starts, it reads any argument other than options as
792 specifying an executable file. This is the same as if the argument was
793 specified by the @samp{-se} option.
796 Many options have both long and short forms; both are shown in the
797 following list. @value{GDBN} also recognizes the long forms if you truncate
798 them, so long as enough of the option is present to be unambiguous.
799 (If you prefer, you can flag option arguments with @samp{--} rather
800 than @samp{-}, though we illustrate the more usual convention.)
803 @item -symbols @var{file}
805 Read symbol table from file @var{file}.
807 @item -exec @var{file}
809 Use file @var{file} as the executable file to execute when
814 appropriate, and for examining pure data in conjunction with a core
819 Read symbol table from file @var{file} and use it as the executable
823 @item -core @var{file}
825 Use file @var{file} as a core dump to examine.
827 @item -c @var{number}
828 Connect to process ID @var{number}, as with the @code{attach} command
829 (unless there is a file in core-dump format named @var{number}, in which
830 case @samp{-c} specifies that file as a core dump to read).
833 @item -command @var{file}
835 Execute @value{GDBN} commands from file @var{file}. @xref{Command
836 Files,, Command files}.
838 @item -directory @var{directory}
839 @itemx -d @var{directory}
840 Add @var{directory} to the path to search for source files.
845 @emph{Warning: this option depends on operating system facilities that are not
846 supported on all systems.}@*
847 If memory-mapped files are available on your system through the @code{mmap}
848 system call, you can use this option
849 to have @value{GDBN} write the symbols from your
850 program into a reusable file in the current directory. If the program you are debugging is
851 called @file{/tmp/fred}, the mapped symbol file is @file{./fred.syms}.
852 Future @value{GDBN} debugging sessions notice the presence of this file,
853 and can quickly map in symbol information from it, rather than reading
854 the symbol table from the executable program.
856 The @file{.syms} file is specific to the host machine where @value{GDBN}
857 is run. It holds an exact image of the internal @value{GDBN} symbol
858 table. It cannot be shared across multiple host platforms.
863 Read each symbol file's entire symbol table immediately, rather than
864 the default, which is to read it incrementally as it is needed.
865 This makes startup slower, but makes future operations faster.
869 The @code{-mapped} and @code{-readnow} options are typically combined in
870 order to build a @file{.syms} file that contains complete symbol
871 information. (@xref{Files,,Commands to specify files}, for information
873 a @file{.syms} file for future use is:
876 gdb -batch -nx -mapped -readnow programname
881 @subsection Choosing modes
883 You can run @value{GDBN} in various alternative modes---for example, in
884 batch mode or quiet mode.
889 Do not execute commands from any initialization files (normally called
890 @file{@value{GDBINIT}}). Normally, the commands in these files are
891 executed after all the command options and arguments have been
892 processed. @xref{Command Files,,Command files}.
896 ``Quiet''. Do not print the introductory and copyright messages. These
897 messages are also suppressed in batch mode.
900 Run in batch mode. Exit with status @code{0} after processing all the
901 command files specified with @samp{-x} (and all commands from
902 initialization files, if not inhibited with @samp{-n}). Exit with
903 nonzero status if an error occurs in executing the @value{GDBN} commands
904 in the command files.
906 Batch mode may be useful for running @value{GDBN} as a filter, for example to
907 download and run a program on another computer; in order to make this
908 more useful, the message
911 Program exited normally.
915 (which is ordinarily issued whenever a program running under @value{GDBN} control
916 terminates) is not issued when running in batch mode.
918 @item -cd @var{directory}
919 Run @value{GDBN} using @var{directory} as its working directory,
920 instead of the current directory.
923 @item -context @var{authentication}
924 When the Energize programming system starts up @value{GDBN}, it uses this
925 option to trigger an alternate mode of interaction.
926 @var{authentication} is a pair of numeric codes that identify @value{GDBN}
927 as a client in the Energize environment. Avoid this option when you run
928 @value{GDBN} directly from the command line. See @ref{Energize,,Using
929 @value{GDBN} with Energize} for more discussion of using @value{GDBN} with Energize.
935 @sc{gnu} Emacs sets this option when it runs @value{GDBN} as a subprocess. It tells @value{GDBN}
936 to output the full file name and line number in a standard,
937 recognizable fashion each time a stack frame is displayed (which
938 includes each time your program stops). This recognizable format looks
939 like two @samp{\032} characters, followed by the file name, line number
940 and character position separated by colons, and a newline. The
941 Emacs-to-@value{GDBN} interface program uses the two @samp{\032} characters as
942 a signal to display the source code for the frame.
947 Set the line speed (baud rate or bits per second) of any serial
948 interface used by @value{GDBN} for remote debugging.
950 @item -tty @var{device}
951 Run using @var{device} for your program's standard input and output.
952 @c FIXME: kingdon thinks there is more to -tty. Investigate.
957 @section Quitting @value{GDBN}
958 @cindex exiting @value{GDBN}
959 @cindex leaving @value{GDBN}
962 @kindex quit @r{[}@var{expression}@r{]}
965 To exit @value{GDBN}, use the @code{quit} command (abbreviated @code{q}), or
966 type an end-of-file character (usually @kbd{C-d}). If you do not supply
967 @var{expression}, @value{GDBN} will terminate normally; otherwise it will
968 terminate using the result of @var{expression} as the error code.
972 An interrupt (often @kbd{C-c}) does not exit from @value{GDBN}, but rather
973 terminates the action of any @value{GDBN} command that is in progress and
974 returns to @value{GDBN} command level. It is safe to type the interrupt
975 character at any time because @value{GDBN} does not allow it to take effect
976 until a time when it is safe.
979 If you have been using @value{GDBN} to control an attached process or
980 device, you can release it with the @code{detach} command
981 (@pxref{Attach, ,Debugging an already-running process}).
985 @section Shell commands
987 If you need to execute occasional shell commands during your
988 debugging session, there is no need to leave or suspend @value{GDBN}; you can
989 just use the @code{shell} command.
994 @item shell @var{command string}
995 Invoke a the standard shell to execute @var{command string}.
997 If it exists, the environment variable @code{SHELL} determines which
998 shell to run. Otherwise @value{GDBN} uses @code{/bin/sh}.
1002 The utility @code{make} is often needed in development environments.
1003 You do not have to use the @code{shell} command for this purpose in
1008 @cindex calling make
1009 @item make @var{make-args}
1010 Execute the @code{make} program with the specified
1011 arguments. This is equivalent to @samp{shell make @var{make-args}}.
1015 @chapter @value{GDBN} Commands
1017 You can abbreviate a @value{GDBN} command to the first few letters of the command
1018 name, if that abbreviation is unambiguous; and you can repeat certain
1019 @value{GDBN} commands by typing just @key{RET}. You can also use the @key{TAB}
1020 key to get @value{GDBN} to fill out the rest of a word in a command (or to
1021 show you the alternatives available, if there is more than one possibility).
1024 * Command Syntax:: How to give commands to @value{GDBN}
1025 * Completion:: Command completion
1026 * Help:: How to ask @value{GDBN} for help
1029 @node Command Syntax
1030 @section Command syntax
1032 A @value{GDBN} command is a single line of input. There is no limit on
1033 how long it can be. It starts with a command name, which is followed by
1034 arguments whose meaning depends on the command name. For example, the
1035 command @code{step} accepts an argument which is the number of times to
1036 step, as in @samp{step 5}. You can also use the @code{step} command
1037 with no arguments. Some command names do not allow any arguments.
1039 @cindex abbreviation
1040 @value{GDBN} command names may always be truncated if that abbreviation is
1041 unambiguous. Other possible command abbreviations are listed in the
1042 documentation for individual commands. In some cases, even ambiguous
1043 abbreviations are allowed; for example, @code{s} is specially defined as
1044 equivalent to @code{step} even though there are other commands whose
1045 names start with @code{s}. You can test abbreviations by using them as
1046 arguments to the @code{help} command.
1048 @cindex repeating commands
1050 A blank line as input to @value{GDBN} (typing just @key{RET}) means to
1051 repeat the previous command. Certain commands (for example, @code{run})
1052 will not repeat this way; these are commands whose unintentional
1053 repetition might cause trouble and which you are unlikely to want to
1056 The @code{list} and @code{x} commands, when you repeat them with
1057 @key{RET}, construct new arguments rather than repeating
1058 exactly as typed. This permits easy scanning of source or memory.
1060 @value{GDBN} can also use @key{RET} in another way: to partition lengthy
1061 output, in a way similar to the common utility @code{more}
1062 (@pxref{Screen Size,,Screen size}). Since it is easy to press one
1063 @key{RET} too many in this situation, @value{GDBN} disables command
1064 repetition after any command that generates this sort of display.
1068 Any text from a @kbd{#} to the end of the line is a comment; it does
1069 nothing. This is useful mainly in command files (@pxref{Command
1070 Files,,Command files}).
1073 @section Command completion
1076 @cindex word completion
1077 @value{GDBN} can fill in the rest of a word in a command for you, if there is
1078 only one possibility; it can also show you what the valid possibilities
1079 are for the next word in a command, at any time. This works for @value{GDBN}
1080 commands, @value{GDBN} subcommands, and the names of symbols in your program.
1082 Press the @key{TAB} key whenever you want @value{GDBN} to fill out the rest
1083 of a word. If there is only one possibility, @value{GDBN} fills in the
1084 word, and waits for you to finish the command (or press @key{RET} to
1085 enter it). For example, if you type
1087 @c FIXME "@key" does not distinguish its argument sufficiently to permit
1088 @c complete accuracy in these examples; space introduced for clarity.
1089 @c If texinfo enhancements make it unnecessary, it would be nice to
1090 @c replace " @key" by "@key" in the following...
1092 (@value{GDBP}) info bre @key{TAB}
1096 @value{GDBN} fills in the rest of the word @samp{breakpoints}, since that is
1097 the only @code{info} subcommand beginning with @samp{bre}:
1100 (@value{GDBP}) info breakpoints
1104 You can either press @key{RET} at this point, to run the @code{info
1105 breakpoints} command, or backspace and enter something else, if
1106 @samp{breakpoints} does not look like the command you expected. (If you
1107 were sure you wanted @code{info breakpoints} in the first place, you
1108 might as well just type @key{RET} immediately after @samp{info bre},
1109 to exploit command abbreviations rather than command completion).
1111 If there is more than one possibility for the next word when you press
1112 @key{TAB}, @value{GDBN} sounds a bell. You can either supply more
1113 characters and try again, or just press @key{TAB} a second time;
1114 @value{GDBN} displays all the possible completions for that word. For
1115 example, you might want to set a breakpoint on a subroutine whose name
1116 begins with @samp{make_}, but when you type @kbd{b make_@key{TAB}} @value{GDBN}
1117 just sounds the bell. Typing @key{TAB} again displays all the
1118 function names in your program that begin with those characters, for
1122 (@value{GDBP}) b make_ @key{TAB}
1123 @exdent @value{GDBN} sounds bell; press @key{TAB} again, to see:
1124 make_a_section_from_file make_environ
1125 make_abs_section make_function_type
1126 make_blockvector make_pointer_type
1127 make_cleanup make_reference_type
1128 make_command make_symbol_completion_list
1129 (@value{GDBP}) b make_
1133 After displaying the available possibilities, @value{GDBN} copies your
1134 partial input (@samp{b make_} in the example) so you can finish the
1137 If you just want to see the list of alternatives in the first place, you
1138 can press @kbd{M-?} rather than pressing @key{TAB} twice. @kbd{M-?}
1139 means @kbd{@key{META} ?}. You can type this
1141 either by holding down a
1142 key designated as the @key{META} shift on your keyboard (if there is
1143 one) while typing @kbd{?}, or
1145 as @key{ESC} followed by @kbd{?}.
1147 @cindex quotes in commands
1148 @cindex completion of quoted strings
1149 Sometimes the string you need, while logically a ``word'', may contain
1150 parentheses or other characters that @value{GDBN} normally excludes from its
1151 notion of a word. To permit word completion to work in this situation,
1152 you may enclose words in @code{'} (single quote marks) in @value{GDBN} commands.
1155 The most likely situation where you might need this is in typing the
1156 name of a C++ function. This is because C++ allows function overloading
1157 (multiple definitions of the same function, distinguished by argument
1158 type). For example, when you want to set a breakpoint you may need to
1159 distinguish whether you mean the version of @code{name} that takes an
1160 @code{int} parameter, @code{name(int)}, or the version that takes a
1161 @code{float} parameter, @code{name(float)}. To use the word-completion
1162 facilities in this situation, type a single quote @code{'} at the
1163 beginning of the function name. This alerts @value{GDBN} that it may need to
1164 consider more information than usual when you press @key{TAB} or
1165 @kbd{M-?} to request word completion:
1168 (@value{GDBP}) b 'bubble( @key{M-?}
1169 bubble(double,double) bubble(int,int)
1170 (@value{GDBP}) b 'bubble(
1173 In some cases, @value{GDBN} can tell that completing a name requires using
1174 quotes. When this happens, @value{GDBN} inserts the quote for you (while
1175 completing as much as it can) if you do not type the quote in the first
1179 (@value{GDBP}) b bub @key{TAB}
1180 @exdent @value{GDBN} alters your input line to the following, and rings a bell:
1181 (@value{GDBP}) b 'bubble(
1185 In general, @value{GDBN} can tell that a quote is needed (and inserts it) if
1186 you have not yet started typing the argument list when you ask for
1187 completion on an overloaded symbol.
1192 @section Getting help
1193 @cindex online documentation
1196 You can always ask @value{GDBN} itself for information on its commands,
1197 using the command @code{help}.
1203 You can use @code{help} (abbreviated @code{h}) with no arguments to
1204 display a short list of named classes of commands:
1208 List of classes of commands:
1210 running -- Running the program
1211 stack -- Examining the stack
1212 data -- Examining data
1213 breakpoints -- Making program stop at certain points
1214 files -- Specifying and examining files
1215 status -- Status inquiries
1216 support -- Support facilities
1217 user-defined -- User-defined commands
1218 aliases -- Aliases of other commands
1219 obscure -- Obscure features
1221 Type "help" followed by a class name for a list of
1222 commands in that class.
1223 Type "help" followed by command name for full
1225 Command name abbreviations are allowed if unambiguous.
1229 @item help @var{class}
1230 Using one of the general help classes as an argument, you can get a
1231 list of the individual commands in that class. For example, here is the
1232 help display for the class @code{status}:
1235 (@value{GDBP}) help status
1240 @c Line break in "show" line falsifies real output, but needed
1241 @c to fit in smallbook page size.
1242 show -- Generic command for showing things set
1244 info -- Generic command for printing status
1246 Type "help" followed by command name for full
1248 Command name abbreviations are allowed if unambiguous.
1252 @item help @var{command}
1253 With a command name as @code{help} argument, @value{GDBN} displays a
1254 short paragraph on how to use that command.
1257 @item complete @var{args}
1258 The @code{complete @var{args}} command lists all the possible completions
1259 for the beginning of a command. Use @var{args} to specify the beginning of the
1260 command you want completed. For example:
1266 @noindent results in:
1274 @noindent This is intended for use by @sc{gnu} Emacs.
1277 In addition to @code{help}, you can use the @value{GDBN} commands @code{info}
1278 and @code{show} to inquire about the state of your program, or the state
1279 of @value{GDBN} itself. Each command supports many topics of inquiry; this
1280 manual introduces each of them in the appropriate context. The listings
1281 under @code{info} and under @code{show} in the Index point to
1282 all the sub-commands. @xref{Index}.
1289 This command (abbreviated @code{i}) is for describing the state of your
1290 program. For example, you can list the arguments given to your program
1291 with @code{info args}, list the registers currently in use with @code{info
1292 registers}, or list the breakpoints you have set with @code{info breakpoints}.
1293 You can get a complete list of the @code{info} sub-commands with
1294 @w{@code{help info}}.
1298 You can assign the result of an expresson to an environment variable with
1299 @code{set}. For example, you can set the @value{GDBN} prompt to a $-sign with
1300 @code{set prompt $}.
1304 In contrast to @code{info}, @code{show} is for describing the state of
1305 @value{GDBN} itself.
1306 You can change most of the things you can @code{show}, by using the
1307 related command @code{set}; for example, you can control what number
1308 system is used for displays with @code{set radix}, or simply inquire
1309 which is currently in use with @code{show radix}.
1312 To display all the settable parameters and their current
1313 values, you can use @code{show} with no arguments; you may also use
1314 @code{info set}. Both commands produce the same display.
1315 @c FIXME: "info set" violates the rule that "info" is for state of
1316 @c FIXME...program. Ck w/ GNU: "info set" to be called something else,
1317 @c FIXME...or change desc of rule---eg "state of prog and debugging session"?
1321 Here are three miscellaneous @code{show} subcommands, all of which are
1322 exceptional in lacking corresponding @code{set} commands:
1325 @kindex show version
1326 @cindex version number
1328 Show what version of @value{GDBN} is running. You should include this
1329 information in @value{GDBN} bug-reports. If multiple versions of @value{GDBN} are in
1330 use at your site, you may occasionally want to determine which version
1331 of @value{GDBN} you are running; as @value{GDBN} evolves, new commands are introduced,
1332 and old ones may wither away. The version number is also announced
1333 when you start @value{GDBN}.
1335 @kindex show copying
1337 Display information about permission for copying @value{GDBN}.
1339 @kindex show warranty
1341 Display the @sc{gnu} ``NO WARRANTY'' statement.
1345 @chapter Running Programs Under @value{GDBN}
1347 When you run a program under @value{GDBN}, you must first generate
1348 debugging information when you compile it.
1350 You may start @value{GDBN} with its arguments, if any, in an environment
1351 of your choice. You may redirect your program's input and output, debug an
1352 already running process, or kill a child process.
1356 * Compilation:: Compiling for debugging
1357 * Starting:: Starting your program
1359 * Arguments:: Your program's arguments
1360 * Environment:: Your program's environment
1361 * Working Directory:: Your program's working directory
1362 * Input/Output:: Your program's input and output
1363 * Attach:: Debugging an already-running process
1364 * Kill Process:: Killing the child process
1365 * Process Information:: Additional process information
1366 * Threads:: Debugging programs with multiple threads
1367 * Processes:: Debugging programs with multiple processes
1372 @section Compiling for debugging
1374 In order to debug a program effectively, you need to generate
1375 debugging information when you compile it. This debugging information
1376 is stored in the object file; it describes the data type of each
1377 variable or function and the correspondence between source line numbers
1378 and addresses in the executable code.
1380 To request debugging information, specify the @samp{-g} option when you run
1383 Many C compilers are unable to handle the @samp{-g} and @samp{-O}
1384 options together. Using those compilers, you cannot generate optimized
1385 executables containing debugging information.
1387 @value{NGCC}, the @sc{gnu} C compiler, supports @samp{-g} with or without
1388 @samp{-O}, making it possible to debug optimized code. We recommend
1389 that you @emph{always} use @samp{-g} whenever you compile a program.
1390 You may think your program is correct, but there is no sense in pushing
1393 @cindex optimized code, debugging
1394 @cindex debugging optimized code
1395 When you debug a program compiled with @samp{-g -O}, remember that the
1396 optimizer is rearranging your code; the debugger shows you what is
1397 really there. Do not be too surprised when the execution path does not
1398 exactly match your source file! An extreme example: if you define a
1399 variable, but never use it, @value{GDBN} never sees that
1400 variable---because the compiler optimizes it out of existence.
1402 Some things do not work as well with @samp{-g -O} as with just
1403 @samp{-g}, particularly on machines with instruction scheduling. If in
1404 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
1405 please report it to us as a bug (including a test case!).
1407 Older versions of the @sc{gnu} C compiler permitted a variant option
1408 @w{@samp{-gg}} for debugging information. @value{GDBN} no longer supports this
1409 format; if your @sc{gnu} C compiler has this option, do not use it.
1413 @section Starting your program
1421 Use the @code{run} command to start your program under @value{GDBN}. You must
1422 first specify the program name
1426 with an argument to @value{GDBN} (@pxref{Invocation, ,Getting In and
1427 Out of @value{GDBN}}), or by using the @code{file} or @code{exec-file}
1428 command (@pxref{Files, ,Commands to specify files}).
1433 If you are running your program in an execution environment that
1434 supports processes, @code{run} creates an inferior process and makes
1435 that process run your program. (In environments without processes,
1436 @code{run} jumps to the start of your program.)
1438 The execution of a program is affected by certain information it
1439 receives from its superior. @value{GDBN} provides ways to specify this
1440 information, which you must do @emph{before} starting your program. (You
1441 can change it after starting your program, but such changes only affect
1442 your program the next time you start it.) This information may be
1443 divided into four categories:
1446 @item The @emph{arguments.}
1447 Specify the arguments to give your program as the arguments of the
1448 @code{run} command. If a shell is available on your target, the shell
1449 is used to pass the arguments, so that you may use normal conventions
1450 (such as wildcard expansion or variable substitution) in describing
1451 the arguments. In Unix systems, you can control which shell is used
1452 with the @code{SHELL} environment variable. @xref{Arguments, ,Your
1453 program's arguments}.
1455 @item The @emph{environment.}
1456 Your program normally inherits its environment from @value{GDBN}, but you can
1457 use the @value{GDBN} commands @code{set environment} and @code{unset
1458 environment} to change parts of the environment that affect
1459 your program. @xref{Environment, ,Your program's environment}.
1461 @item The @emph{working directory.}
1462 Your program inherits its working directory from @value{GDBN}. You can set
1463 the @value{GDBN} working directory with the @code{cd} command in @value{GDBN}.
1464 @xref{Working Directory, ,Your program's working directory}.
1466 @item The @emph{standard input and output.}
1467 Your program normally uses the same device for standard input and
1468 standard output as @value{GDBN} is using. You can redirect input and output
1469 in the @code{run} command line, or you can use the @code{tty} command to
1470 set a different device for your program.
1471 @xref{Input/Output, ,Your program's input and output}.
1474 @emph{Warning:} While input and output redirection work, you cannot use
1475 pipes to pass the output of the program you are debugging to another
1476 program; if you attempt this, @value{GDBN} is likely to wind up debugging the
1481 When you issue the @code{run} command, your program begins to execute
1482 immediately. @xref{Stopping, ,Stopping and continuing}, for discussion
1483 of how to arrange for your program to stop. Once your program has
1484 stopped, you may call functions in your program, using the @code{print}
1485 or @code{call} commands. @xref{Data, ,Examining Data}.
1487 If the modification time of your symbol file has changed since the last
1488 time @value{GDBN} read its symbols, @value{GDBN} discards its symbol
1489 table, and reads it again. When it does this, @value{GDBN} tries to retain
1490 your current breakpoints.
1494 @section Your program's arguments
1496 @cindex arguments (to your program)
1497 The arguments to your program can be specified by the arguments of the
1498 @code{run} command. They are passed to a shell, which expands wildcard
1499 characters and performs redirection of I/O, and thence to your program.
1500 Your @code{SHELL} environment variable (if it exists) specifies what
1501 shell @value{GDBN} uses. If you do not define @code{SHELL},
1502 @value{GDBN} uses @code{/bin/sh}.
1504 @code{run} with no arguments uses the same arguments used by the previous
1505 @code{run}, or those set by the @code{set args} command.
1510 Specify the arguments to be used the next time your program is run. If
1511 @code{set args} has no arguments, @code{run} executes your program
1512 with no arguments. Once you have run your program with arguments,
1513 using @code{set args} before the next @code{run} is the only way to run
1514 it again without arguments.
1518 Show the arguments to give your program when it is started.
1522 @section Your program's environment
1524 @cindex environment (of your program)
1525 The @dfn{environment} consists of a set of environment variables and
1526 their values. Environment variables conventionally record such things as
1527 your user name, your home directory, your terminal type, and your search
1528 path for programs to run. Usually you set up environment variables with
1529 the shell and they are inherited by all the other programs you run. When
1530 debugging, it can be useful to try running your program with a modified
1531 environment without having to start @value{GDBN} over again.
1535 @item path @var{directory}
1536 Add @var{directory} to the front of the @code{PATH} environment variable
1537 (the search path for executables), for both @value{GDBN} and your program.
1538 You may specify several directory names, separated by @samp{:} or
1539 whitespace. If @var{directory} is already in the path, it is moved to
1540 the front, so it is searched sooner.
1542 You can use the string @samp{$cwd} to refer to whatever is the current
1543 working directory at the time @value{GDBN} searches the path. If you
1544 use @samp{.} instead, it refers to the directory where you executed the
1545 @code{path} command. @value{GDBN} replaces @samp{.} in the
1546 @var{directory} argument (with the current path) before adding
1547 @var{directory} to the search path.
1548 @c 'path' is explicitly nonrepeatable, but RMS points out it is silly to
1549 @c document that, since repeating it would be a no-op.
1553 Display the list of search paths for executables (the @code{PATH}
1554 environment variable).
1556 @kindex show environment
1557 @item show environment @r{[}@var{varname}@r{]}
1558 Print the value of environment variable @var{varname} to be given to
1559 your program when it starts. If you do not supply @var{varname},
1560 print the names and values of all environment variables to be given to
1561 your program. You can abbreviate @code{environment} as @code{env}.
1563 @kindex set environment
1564 @item set environment @var{varname} @r{[}=@r{]} @var{value}
1565 Set environment variable @var{varname} to @var{value}. The value
1566 changes for your program only, not for @value{GDBN} itself. @var{value} may
1567 be any string; the values of environment variables are just strings, and
1568 any interpretation is supplied by your program itself. The @var{value}
1569 parameter is optional; if it is eliminated, the variable is set to a
1571 @c "any string" here does not include leading, trailing
1572 @c blanks. Gnu asks: does anyone care?
1574 For example, this command:
1581 tells a Unix program, when subsequently run, that its user is named
1582 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
1583 are not actually required.)
1585 @kindex unset environment
1586 @item unset environment @var{varname}
1587 Remove variable @var{varname} from the environment to be passed to your
1588 program. This is different from @samp{set env @var{varname} =};
1589 @code{unset environment} removes the variable from the environment,
1590 rather than assigning it an empty value.
1593 @emph{Warning:} @value{GDBN} runs your program using the shell indicated
1594 by your @code{SHELL} environment variable if it exists (or
1595 @code{/bin/sh} if not). If your @code{SHELL} variable names a shell
1596 that runs an initialization file---such as @file{.cshrc} for C-shell, or
1597 @file{.bashrc} for BASH---any variables you set in that file affect
1598 your program. You may wish to move setting of environment variables to
1599 files that are only run when you sign on, such as @file{.login} or
1602 @node Working Directory
1603 @section Your program's working directory
1605 @cindex working directory (of your program)
1606 Each time you start your program with @code{run}, it inherits its
1607 working directory from the current working directory of @value{GDBN}.
1608 The @value{GDBN} working directory is initially whatever it inherited
1609 from its parent process (typically the shell), but you can specify a new
1610 working directory in @value{GDBN} with the @code{cd} command.
1612 The @value{GDBN} working directory also serves as a default for the commands
1613 that specify files for @value{GDBN} to operate on. @xref{Files, ,Commands to
1618 @item cd @var{directory}
1619 Set the @value{GDBN} working directory to @var{directory}.
1623 Print the @value{GDBN} working directory.
1627 @section Your program's input and output
1632 By default, the program you run under @value{GDBN} does input and output to
1633 the same terminal that @value{GDBN} uses. @value{GDBN} switches the terminal
1634 to its own terminal modes to interact with you, but it records the terminal
1635 modes your program was using and switches back to them when you continue
1636 running your program.
1639 @kindex info terminal
1641 Displays information recorded by @value{GDBN} about the terminal modes your
1645 You can redirect your program's input and/or output using shell
1646 redirection with the @code{run} command. For example,
1653 starts your program, diverting its output to the file @file{outfile}.
1656 @cindex controlling terminal
1657 Another way to specify where your program should do input and output is
1658 with the @code{tty} command. This command accepts a file name as
1659 argument, and causes this file to be the default for future @code{run}
1660 commands. It also resets the controlling terminal for the child
1661 process, for future @code{run} commands. For example,
1668 directs that processes started with subsequent @code{run} commands
1669 default to do input and output on the terminal @file{/dev/ttyb} and have
1670 that as their controlling terminal.
1672 An explicit redirection in @code{run} overrides the @code{tty} command's
1673 effect on the input/output device, but not its effect on the controlling
1676 When you use the @code{tty} command or redirect input in the @code{run}
1677 command, only the input @emph{for your program} is affected. The input
1678 for @value{GDBN} still comes from your terminal.
1681 @section Debugging an already-running process
1686 @item attach @var{process-id}
1687 This command attaches to a running process---one that was started
1688 outside @value{GDBN}. (@code{info files} shows your active
1689 targets.) The command takes as argument a process ID. The usual way to
1690 find out the process-id of a Unix process is with the @code{ps} utility,
1691 or with the @samp{jobs -l} shell command.
1693 @code{attach} does not repeat if you press @key{RET} a second time after
1694 executing the command.
1697 To use @code{attach}, your program must be running in an environment
1698 which supports processes; for example, @code{attach} does not work for
1699 programs on bare-board targets that lack an operating system. You must
1700 also have permission to send the process a signal.
1702 When using @code{attach}, you should first use the @code{file} command
1703 to specify the program running in the process and load its symbol table.
1704 @xref{Files, ,Commands to Specify Files}.
1706 The first thing @value{GDBN} does after arranging to debug the specified
1707 process is to stop it. You can examine and modify an attached process
1708 with all the @value{GDBN} commands that are ordinarily available when you start
1709 processes with @code{run}. You can insert breakpoints; you can step and
1710 continue; you can modify storage. If you would rather the process
1711 continue running, you may use the @code{continue} command after
1712 attaching @value{GDBN} to the process.
1717 When you have finished debugging the attached process, you can use the
1718 @code{detach} command to release it from @value{GDBN} control. Detaching
1719 the process continues its execution. After the @code{detach} command,
1720 that process and @value{GDBN} become completely independent once more, and you
1721 are ready to @code{attach} another process or start one with @code{run}.
1722 @code{detach} does not repeat if you press @key{RET} again after
1723 executing the command.
1726 If you exit @value{GDBN} or use the @code{run} command while you have an
1727 attached process, you kill that process. By default, @value{GDBN} asks
1728 for confirmation if you try to do either of these things; you can
1729 control whether or not you need to confirm by using the @code{set
1730 confirm} command (@pxref{Messages/Warnings, ,Optional warnings and
1735 @section Killing the child process
1740 Kill the child process in which your program is running under @value{GDBN}.
1743 This command is useful if you wish to debug a core dump instead of a
1744 running process. @value{GDBN} ignores any core dump file while your program
1748 On some operating systems, a program cannot be executed outside @value{GDBN}
1749 while you have breakpoints set on it inside @value{GDBN}. You can use the
1750 @code{kill} command in this situation to permit running your program
1751 outside the debugger.
1753 The @code{kill} command is also useful if you wish to recompile and
1754 relink your program, since on many systems it is impossible to modify an
1755 executable file while it is running in a process. In this case, when you
1756 next type @code{run}, @value{GDBN} notices that the file has changed, and
1757 reads the symbol table again (while trying to preserve your current
1758 breakpoint settings).
1760 @node Process Information
1761 @section Additional process information
1764 @cindex process image
1765 Some operating systems provide a facility called @samp{/proc} that can
1766 be used to examine the image of a running process using file-system
1767 subroutines. If @value{GDBN} is configured for an operating system with this
1768 facility, the command @code{info proc} is available to report on several
1769 kinds of information about the process running your program.
1770 @code{info proc} works only on SVR4 systems that support @code{procfs}.
1775 Summarize available information about the process.
1777 @kindex info proc mappings
1778 @item info proc mappings
1779 Report on the address ranges accessible in the program, with information
1780 on whether your program may read, write, or execute each range.
1782 @kindex info proc times
1783 @item info proc times
1784 Starting time, user CPU time, and system CPU time for your program and
1787 @kindex info proc id
1789 Report on the process IDs related to your program: its own process ID,
1790 the ID of its parent, the process group ID, and the session ID.
1792 @kindex info proc status
1793 @item info proc status
1794 General information on the state of the process. If the process is
1795 stopped, this report includes the reason for stopping, and any signal
1799 Show all the above information about the process.
1803 @section Debugging programs with multiple threads
1805 @cindex threads of execution
1806 @cindex multiple threads
1807 @cindex switching threads
1808 In some operating systems, a single program may have more than one
1809 @dfn{thread} of execution. The precise semantics of threads differ from
1810 one operating system to another, but in general the threads of a single
1811 program are akin to multiple processes---except that they share one
1812 address space (that is, they can all examine and modify the same
1813 variables). On the other hand, each thread has its own registers and
1814 execution stack, and perhaps private memory.
1816 @value{GDBN} provides these facilities for debugging multi-thread
1820 @item automatic notification of new threads
1821 @item @samp{thread @var{threadno}}, a command to switch among threads
1822 @item @samp{info threads}, a command to inquire about existing threads
1823 @item @samp{thread apply [@var{threadno}] [@var{all}] @var{args}},
1824 a command to apply a command to a list of threads
1825 @item thread-specific breakpoints
1829 @emph{Warning:} These facilities are not yet available on every
1830 @value{GDBN} configuration where the operating system supports threads.
1831 If your @value{GDBN} does not support threads, these commands have no
1832 effect. For example, a system without thread support shows no output
1833 from @samp{info threads}, and always rejects the @code{thread} command,
1837 (@value{GDBP}) info threads
1838 (@value{GDBP}) thread 1
1839 Thread ID 1 not known. Use the "info threads" command to
1840 see the IDs of currently known threads.
1842 @c FIXME to implementors: how hard would it be to say "sorry, this GDB
1843 @c doesn't support threads"?
1846 @cindex focus of debugging
1847 @cindex current thread
1848 The @value{GDBN} thread debugging facility allows you to observe all
1849 threads while your program runs---but whenever @value{GDBN} takes
1850 control, one thread in particular is always the focus of debugging.
1851 This thread is called the @dfn{current thread}. Debugging commands show
1852 program information from the perspective of the current thread.
1854 @kindex New @var{systag}
1855 @cindex thread identifier (system)
1856 @c FIXME-implementors!! It would be more helpful if the [New...] message
1857 @c included GDB's numeric thread handle, so you could just go to that
1858 @c thread without first checking `info threads'.
1859 Whenever @value{GDBN} detects a new thread in your program, it displays
1860 the target system's identification for the thread with a message in the
1861 form @samp{[New @var{systag}]}. @var{systag} is a thread identifier
1862 whose form varies depending on the particular system. For example, on
1863 LynxOS, you might see
1866 [New process 35 thread 27]
1870 when @value{GDBN} notices a new thread. In contrast, on an SGI system,
1871 the @var{systag} is simply something like @samp{process 368}, with no
1874 @c FIXME!! (1) Does the [New...] message appear even for the very first
1875 @c thread of a program, or does it only appear for the
1876 @c second---i.e., when it becomes obvious we have a multithread
1878 @c (2) *Is* there necessarily a first thread always? Or do some
1879 @c multithread systems permit starting a program with multiple
1880 @c threads ab initio?
1882 @cindex thread number
1883 @cindex thread identifier (GDB)
1884 For debugging purposes, @value{GDBN} associates its own thread
1885 number---always a single integer---with each thread in your program.
1888 @kindex info threads
1890 Display a summary of all threads currently in your
1891 program. @value{GDBN} displays for each thread (in this order):
1894 @item the thread number assigned by @value{GDBN}
1896 @item the target system's thread identifier (@var{systag})
1898 @item the current stack frame summary for that thread
1902 An asterisk @samp{*} to the left of the @value{GDBN} thread number
1903 indicates the current thread.
1907 @c end table here to get a little more width for example
1910 (@value{GDBP}) info threads
1911 3 process 35 thread 27 0x34e5 in sigpause ()
1912 2 process 35 thread 23 0x34e5 in sigpause ()
1913 * 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
1918 @kindex thread @var{threadno}
1919 @item thread @var{threadno}
1920 Make thread number @var{threadno} the current thread. The command
1921 argument @var{threadno} is the internal @value{GDBN} thread number, as
1922 shown in the first field of the @samp{info threads} display.
1923 @value{GDBN} responds by displaying the system identifier of the thread
1924 you selected, and its current stack frame summary:
1927 @c FIXME!! This example made up; find a @value{GDBN} w/threads and get real one
1928 (@value{GDBP}) thread 2
1929 [Switching to process 35 thread 23]
1930 0x34e5 in sigpause ()
1934 As with the @samp{[New @dots{}]} message, the form of the text after
1935 @samp{Switching to} depends on your system's conventions for identifying
1938 @kindex thread apply
1939 @item thread apply [@var{threadno}] [@var{all}] @var{args}
1940 The @code{thread apply} command allows you to apply a command to one or
1941 more threads. Specify the numbers of the threads that you want affected
1942 with the command argument @var{threadno}. @var{threadno} is the internal
1943 @value{GDBN} thread number, as shown in the first field of the @samp{info
1944 threads} display. To apply a command to all threads, use
1945 @code{thread apply all} @var{args}.
1948 @cindex automatic thread selection
1949 @cindex switching threads automatically
1950 @cindex threads, automatic switching
1951 Whenever @value{GDBN} stops your program, due to a breakpoint or a
1952 signal, it automatically selects the thread where that breakpoint or
1953 signal happened. @value{GDBN} alerts you to the context switch with a
1954 message of the form @samp{[Switching to @var{systag}]} to identify the
1957 @xref{Thread Stops,,Stopping and starting multi-thread programs}, for
1958 more information about how @value{GDBN} behaves when you stop and start
1959 programs with multiple threads.
1961 @xref{Set Watchpoints,,Setting watchpoints}, for information about
1962 watchpoints in programs with multiple threads.
1966 @section Debugging programs with multiple processes
1968 @cindex fork, debugging programs which call
1969 @cindex multiple processes
1970 @cindex processes, multiple
1971 @value{GDBN} has no special support for debugging programs which create
1972 additional processes using the @code{fork} function. When a program
1973 forks, @value{GDBN} will continue to debug the parent process and the
1974 child process will run unimpeded. If you have set a breakpoint in any
1975 code which the child then executes, the child will get a @code{SIGTRAP}
1976 signal which (unless it catches the signal) will cause it to terminate.
1978 However, if you want to debug the child process there is a workaround
1979 which isn't too painful. Put a call to @code{sleep} in the code which
1980 the child process executes after the fork. It may be useful to sleep
1981 only if a certain environment variable is set, or a certain file exists,
1982 so that the delay need not occur when you don't want to run @value{GDBN}
1983 on the child. While the child is sleeping, use the @code{ps} program to
1984 get its process ID. Then tell @value{GDBN} (a new invocation of
1985 @value{GDBN} if you are also debugging the parent process) to attach to
1986 the child process (see @ref{Attach}). From that point on you can debug
1987 the child process just like any other process which you attached to.
1990 @chapter Stopping and Continuing
1992 The principal purposes of using a debugger are so that you can stop your
1993 program before it terminates; or so that, if your program runs into
1994 trouble, you can investigate and find out why.
1996 Inside @value{GDBN}, your program may stop for any of several reasons, such
2001 a breakpoint, or reaching a new line after a @value{GDBN}
2002 command such as @code{step}. You may then examine and change
2003 variables, set new breakpoints or remove old ones, and then continue
2004 execution. Usually, the messages shown by @value{GDBN} provide ample
2005 explanation of the status of your program---but you can also explicitly
2006 request this information at any time.
2009 @kindex info program
2011 Display information about the status of your program: whether it is
2021 * Breakpoints:: Breakpoints, watchpoints, and exceptions
2024 * Breakpoints:: Breakpoints and watchpoints
2026 @c Remnant makeinfo bug requires blank line after *successful* end-if in menu:
2028 * Continuing and Stepping:: Resuming execution
2033 * Thread Stops:: Stopping and starting multi-thread programs
2037 @c makeinfo node-defaulting requires adjacency of @node and sectioning cmds
2038 @c ...hence distribute @node Breakpoints over two possible @if expansions.
2042 @section Breakpoints, watchpoints, and exceptions
2046 @section Breakpoints and watchpoints
2050 A @dfn{breakpoint} makes your program stop whenever a certain point in
2051 the program is reached. For each breakpoint, you can add
2052 conditions to control in finer detail whether your program stops.
2053 You can set breakpoints with the @code{break} command and its variants
2054 (@pxref{Set Breaks, ,Setting breakpoints}), to specify the place where
2055 your program should stop by line number, function name or exact address
2058 In languages with exception handling (such as @sc{gnu} C++), you can also set
2059 breakpoints where an exception is raised (@pxref{Exception Handling,,
2060 Breakpoints and exceptions}).
2063 In SunOS 4.x, SVR4, and Alpha OSF/1 configurations, you can now set
2064 breakpoints in shared libraries before the executable is run.
2067 @cindex memory tracing
2068 @cindex breakpoint on memory address
2069 @cindex breakpoint on variable modification
2070 A @dfn{watchpoint} is a special breakpoint that stops your program
2071 when the value of an expression changes. You must use a different
2072 command to set watchpoints (@pxref{Set Watchpoints, ,Setting
2073 watchpoints}), but aside from that, you can manage a watchpoint like
2074 any other breakpoint: you enable, disable, and delete both breakpoints
2075 and watchpoints using the same commands.
2077 You can arrange to have values from your program displayed automatically
2078 whenever @value{GDBN} stops at a breakpoint. @xref{Auto Display,,
2081 @cindex breakpoint numbers
2082 @cindex numbers for breakpoints
2083 @value{GDBN} assigns a number to each breakpoint or watchpoint when you
2084 create it; these numbers are successive integers starting with one. In
2085 many of the commands for controlling various features of breakpoints you
2086 use the breakpoint number to say which breakpoint you want to change.
2087 Each breakpoint may be @dfn{enabled} or @dfn{disabled}; if disabled, it has
2088 no effect on your program until you enable it again.
2091 * Set Breaks:: Setting breakpoints
2092 * Set Watchpoints:: Setting watchpoints
2094 * Exception Handling:: Breakpoints and exceptions
2097 * Delete Breaks:: Deleting breakpoints
2098 * Disabling:: Disabling breakpoints
2099 * Conditions:: Break conditions
2100 * Break Commands:: Breakpoint command lists
2102 * Breakpoint Menus:: Breakpoint menus
2104 @c @ifclear BARETARGET
2105 @c * Error in Breakpoints:: ``Cannot insert breakpoints''
2110 @subsection Setting breakpoints
2112 @c FIXME LMB what does GDB do if no code on line of breakpt?
2113 @c consider in particular declaration with/without initialization.
2115 @c FIXME 2 is there stuff on this already? break at fun start, already init?
2120 @cindex latest breakpoint
2121 Breakpoints are set with the @code{break} command (abbreviated
2122 @code{b}). The debugger convenience variable @samp{$bpnum} records the
2123 number of the breakpoints you've set most recently; see @ref{Convenience
2124 Vars,, Convenience variables}, for a discussion of what you can do with
2125 convenience variables.
2127 You have several ways to say where the breakpoint should go.
2130 @item break @var{function}
2131 Set a breakpoint at entry to function @var{function}.
2133 When using source languages that permit overloading of symbols, such as
2134 C++, @var{function} may refer to more than one possible place to break.
2135 @xref{Breakpoint Menus,,Breakpoint menus}, for a discussion of that situation.
2138 @item break +@var{offset}
2139 @itemx break -@var{offset}
2140 Set a breakpoint some number of lines forward or back from the position
2141 at which execution stopped in the currently selected frame.
2143 @item break @var{linenum}
2144 Set a breakpoint at line @var{linenum} in the current source file.
2145 That file is the last file whose source text was printed. This
2146 breakpoint stops your program just before it executes any of the
2149 @item break @var{filename}:@var{linenum}
2150 Set a breakpoint at line @var{linenum} in source file @var{filename}.
2152 @item break @var{filename}:@var{function}
2153 Set a breakpoint at entry to function @var{function} found in file
2154 @var{filename}. Specifying a file name as well as a function name is
2155 superfluous except when multiple files contain similarly named
2158 @item break *@var{address}
2159 Set a breakpoint at address @var{address}. You can use this to set
2160 breakpoints in parts of your program which do not have debugging
2161 information or source files.
2164 When called without any arguments, @code{break} sets a breakpoint at
2165 the next instruction to be executed in the selected stack frame
2166 (@pxref{Stack, ,Examining the Stack}). In any selected frame but the
2167 innermost, this makes your program stop as soon as control
2168 returns to that frame. This is similar to the effect of a
2169 @code{finish} command in the frame inside the selected frame---except
2170 that @code{finish} does not leave an active breakpoint. If you use
2171 @code{break} without an argument in the innermost frame, @value{GDBN} stops
2172 the next time it reaches the current location; this may be useful
2175 @value{GDBN} normally ignores breakpoints when it resumes execution, until at
2176 least one instruction has been executed. If it did not do this, you
2177 would be unable to proceed past a breakpoint without first disabling the
2178 breakpoint. This rule applies whether or not the breakpoint already
2179 existed when your program stopped.
2181 @item break @dots{} if @var{cond}
2182 Set a breakpoint with condition @var{cond}; evaluate the expression
2183 @var{cond} each time the breakpoint is reached, and stop only if the
2184 value is nonzero---that is, if @var{cond} evaluates as true.
2185 @samp{@dots{}} stands for one of the possible arguments described
2186 above (or no argument) specifying where to break. @xref{Conditions,
2187 ,Break conditions}, for more information on breakpoint conditions.
2190 @item tbreak @var{args}
2191 Set a breakpoint enabled only for one stop. @var{args} are the
2192 same as for the @code{break} command, and the breakpoint is set in the same
2193 way, but the breakpoint is automatically deleted after the first time your
2194 program stops there. @xref{Disabling, ,Disabling breakpoints}.
2197 @item hbreak @var{args}
2198 Set a hardware-assisted breakpoint. @var{args} are the same as for the
2199 @code{break} command and the breakpoint is set in the same way, but the
2200 breakpoint requires hardware support and some target hardware may not
2201 have this support. The main purpose of this is EPROM/ROM code
2202 debugging, so you can set a breakpoint at an instruction without
2203 changing the instruction. This can be used with the new trap-generation
2204 provided by SPARClite DSU. DSU will generate traps when a program accesses
2205 some date or instruction address that is assigned to the debug registers.
2206 However the hardware breakpoint registers can only take two data breakpoints,
2207 and @value{GDBN} will reject this command if more than two are used.
2208 Delete or disable usused hardware breakpoints before setting
2209 new ones. @xref{Conditions, ,Break conditions}.
2212 @item thbreak @var{args}
2213 Set a hardware-assisted breakpoint enabled only for one stop. @var{args}
2214 are the same as for the @code{hbreak} command and the breakpoint is set in
2215 the same way. However, like the @code{tbreak} command,
2216 the breakpoint is automatically deleted after the
2217 first time your program stops there. Also, like the @code{hbreak}
2218 command, the breakpoint requires hardware support and some target hardware
2219 may not have this support. @xref{Disabling, ,Disabling breakpoints}.
2220 Also @xref{Conditions, ,Break conditions}.
2223 @cindex regular expression
2224 @item rbreak @var{regex}
2225 @c FIXME what kind of regexp?
2226 Set breakpoints on all functions matching the regular expression
2227 @var{regex}. This command
2228 sets an unconditional breakpoint on all matches, printing a list of all
2229 breakpoints it set. Once these breakpoints are set, they are treated
2230 just like the breakpoints set with the @code{break} command. You can
2231 delete them, disable them, or make them conditional the same way as any
2235 When debugging C++ programs, @code{rbreak} is useful for setting
2236 breakpoints on overloaded functions that are not members of any special
2240 @kindex info breakpoints
2241 @cindex @code{$_} and @code{info breakpoints}
2242 @item info breakpoints @r{[}@var{n}@r{]}
2243 @itemx info break @r{[}@var{n}@r{]}
2244 @itemx info watchpoints @r{[}@var{n}@r{]}
2245 Print a table of all breakpoints and watchpoints set and not
2246 deleted, with the following columns for each breakpoint:
2249 @item Breakpoint Numbers
2251 Breakpoint or watchpoint.
2253 Whether the breakpoint is marked to be disabled or deleted when hit.
2254 @item Enabled or Disabled
2255 Enabled breakpoints are marked with @samp{y}. @samp{n} marks breakpoints
2256 that are not enabled.
2258 Where the breakpoint is in your program, as a memory address
2260 Where the breakpoint is in the source for your program, as a file and
2265 If a breakpoint is conditional, @code{info break} shows the condition on
2266 the line following the affected breakpoint; breakpoint commands, if any,
2267 are listed after that.
2270 @code{info break} with a breakpoint
2271 number @var{n} as argument lists only that breakpoint. The
2272 convenience variable @code{$_} and the default examining-address for
2273 the @code{x} command are set to the address of the last breakpoint
2274 listed (@pxref{Memory, ,Examining memory}).
2277 @code{info break} now displays a count of the number of times the
2278 breakpoint has been hit. This is especially useful in conjunction with
2279 the @code{ignore} command. You can ignore a large number of breakpoint
2280 hits, look at the breakpoint info to see how many times the
2281 breakpoint was hit, and then run again, ignoring one less than that
2282 number. This will get you quickly to the last hit of that breakpoint.
2285 @value{GDBN} allows you to set any number of breakpoints at the same place in
2286 your program. There is nothing silly or meaningless about this. When
2287 the breakpoints are conditional, this is even useful
2288 (@pxref{Conditions, ,Break conditions}).
2290 @cindex negative breakpoint numbers
2291 @cindex internal @value{GDBN} breakpoints
2292 @value{GDBN} itself sometimes sets breakpoints in your program for special
2293 purposes, such as proper handling of @code{longjmp} (in C programs).
2294 These internal breakpoints are assigned negative numbers, starting with
2295 @code{-1}; @samp{info breakpoints} does not display them.
2297 You can see these breakpoints with the @value{GDBN} maintenance command
2298 @samp{maint info breakpoints}.
2301 @kindex maint info breakpoints
2302 @item maint info breakpoints
2303 Using the same format as @samp{info breakpoints}, display both the
2304 breakpoints you've set explicitly, and those @value{GDBN} is using for
2305 internal purposes. Internal breakpoints are shown with negative
2306 breakpoint numbers. The type column identifies what kind of breakpoint
2311 Normal, explicitly set breakpoint.
2314 Normal, explicitly set watchpoint.
2317 Internal breakpoint, used to handle correctly stepping through
2318 @code{longjmp} calls.
2320 @item longjmp resume
2321 Internal breakpoint at the target of a @code{longjmp}.
2324 Temporary internal breakpoint used by the @value{GDBN} @code{until} command.
2327 Temporary internal breakpoint used by the @value{GDBN} @code{finish} command.
2333 @node Set Watchpoints
2334 @subsection Setting watchpoints
2335 @cindex setting watchpoints
2337 You can use a watchpoint to stop execution whenever the value of an
2338 expression changes, without having to predict a particular place
2339 where this may happen.
2341 Watchpoints currently execute two orders of magnitude more slowly than
2342 other breakpoints, but this can be well worth it to catch errors where
2343 you have no clue what part of your program is the culprit.
2345 @c FIXME - did Stan mean to @ignore this out?
2347 Some processors provide special hardware to support watchpoint
2348 evaluation; @value{GDBN} will use such hardware if it is available,
2349 and if the support code has been added for that configuration.
2354 @item watch @var{expr}
2355 Set a watchpoint for an expression. @value{GDBN} will break when @var{expr}
2356 is written into by the program and its value changes.
2357 This can be used with the new trap-generation provided by
2358 SPARClite DSU. DSU will generate traps when a program accesses
2359 some date or instruction address that is assigned to the debug registers.
2360 For the data addresses, DSU facilitates the @code{watch} command.
2361 However the hardware breakpoint registers can only take two data watchpoints,
2362 and both watchpoints must be the same kind. For example, you can set two
2363 watchpoints with @code{watch} commands, two with @code{rwatch}
2364 commands, @strong{or} two with @code{awatch} commands, but you cannot set one
2365 watchpoint with one command and the other with a different command.
2366 @value{GBDN} will reject the command if you try to mix watchpoints.
2367 Delete or disable unused watchpoint commands before setting new ones.
2370 @item rwatch @var{expr}
2371 Set a watchpoint that will break when watch @var{args} is read by the program.
2372 If you use both watchpoints, both must be set with the @code{rwatch}
2376 @item awatch @var{expr}
2377 Set a watchpoint that will break when @var{args} is read and written into
2378 by the program. If you use both watchpoints, both must be set with the
2379 @code{awatch} command.
2381 @kindex info watchpoints
2382 @item info watchpoints
2383 This command prints a list of watchpoints and breakpoints; it is the
2384 same as @code{info break}.
2389 @cindex watchpoints and threads
2390 @cindex threads and watchpoints
2391 @emph{Warning:} in multi-thread programs, watchpoints have only limited
2392 usefulness. With the current watchpoint implementation, @value{GDBN}
2393 can only watch the value of an expression @emph{in a single thread}. If
2394 you are confident that the expression can only change due to the current
2395 thread's activity (and if you are also confident that no other thread
2396 can become current), then you can use watchpoints as usual. However,
2397 @value{GDBN} may not notice when a non-current thread's activity changes
2403 @node Exception Handling
2404 @subsection Breakpoints and exceptions
2405 @cindex exception handlers
2407 Some languages, such as @sc{gnu} C++, implement exception handling. You can
2408 use @value{GDBN} to examine what caused your program to raise an exception,
2409 and to list the exceptions your program is prepared to handle at a
2410 given point in time.
2414 @item catch @var{exceptions}
2415 You can set breakpoints at active exception handlers by using the
2416 @code{catch} command. @var{exceptions} is a list of names of exceptions
2420 You can use @code{info catch} to list active exception handlers.
2421 @xref{Frame Info, ,Information about a frame}.
2423 There are currently some limitations to exception handling in @value{GDBN}:
2427 If you call a function interactively, @value{GDBN} normally returns
2428 control to you when the function has finished executing. If the call
2429 raises an exception, however, the call may bypass the mechanism that
2430 returns control to you and cause your program to simply continue
2431 running until it hits a breakpoint, catches a signal that @value{GDBN} is
2432 listening for, or exits.
2435 You cannot raise an exception interactively.
2438 You cannot install an exception handler interactively.
2441 @cindex raise exceptions
2442 Sometimes @code{catch} is not the best way to debug exception handling:
2443 if you need to know exactly where an exception is raised, it is better to
2444 stop @emph{before} the exception handler is called, since that way you
2445 can see the stack before any unwinding takes place. If you set a
2446 breakpoint in an exception handler instead, it may not be easy to find
2447 out where the exception was raised.
2449 To stop just before an exception handler is called, you need some
2450 knowledge of the implementation. In the case of @sc{gnu} C++, exceptions are
2451 raised by calling a library function named @code{__raise_exception}
2452 which has the following ANSI C interface:
2455 /* @var{addr} is where the exception identifier is stored.
2456 ID is the exception identifier. */
2457 void __raise_exception (void **@var{addr}, void *@var{id});
2461 To make the debugger catch all exceptions before any stack
2462 unwinding takes place, set a breakpoint on @code{__raise_exception}
2463 (@pxref{Breakpoints, ,Breakpoints; watchpoints; and exceptions}).
2465 With a conditional breakpoint (@pxref{Conditions, ,Break conditions})
2466 that depends on the value of @var{id}, you can stop your program when
2467 a specific exception is raised. You can use multiple conditional
2468 breakpoints to stop your program when any of a number of exceptions are
2473 @subsection Deleting breakpoints
2475 @cindex clearing breakpoints, watchpoints
2476 @cindex deleting breakpoints, watchpoints
2477 It is often necessary to eliminate a breakpoint or watchpoint once it
2478 has done its job and you no longer want your program to stop there. This
2479 is called @dfn{deleting} the breakpoint. A breakpoint that has been
2480 deleted no longer exists; it is forgotten.
2482 With the @code{clear} command you can delete breakpoints according to
2483 where they are in your program. With the @code{delete} command you can
2484 delete individual breakpoints or watchpoints by specifying their
2487 It is not necessary to delete a breakpoint to proceed past it. @value{GDBN}
2488 automatically ignores breakpoints on the first instruction to be executed
2489 when you continue execution without changing the execution address.
2494 Delete any breakpoints at the next instruction to be executed in the
2495 selected stack frame (@pxref{Selection, ,Selecting a frame}). When
2496 the innermost frame is selected, this is a good way to delete a
2497 breakpoint where your program just stopped.
2499 @item clear @var{function}
2500 @itemx clear @var{filename}:@var{function}
2501 Delete any breakpoints set at entry to the function @var{function}.
2503 @item clear @var{linenum}
2504 @itemx clear @var{filename}:@var{linenum}
2505 Delete any breakpoints set at or within the code of the specified line.
2507 @cindex delete breakpoints
2510 @item delete @r{[}breakpoints@r{]} @r{[}@var{bnums}@dots{}@r{]}
2511 Delete the breakpoints or watchpoints of the numbers specified as
2512 arguments. If no argument is specified, delete all breakpoints (@value{GDBN}
2513 asks confirmation, unless you have @code{set confirm off}). You
2514 can abbreviate this command as @code{d}.
2518 @subsection Disabling breakpoints
2520 @kindex disable breakpoints
2521 @kindex enable breakpoints
2522 Rather than deleting a breakpoint or watchpoint, you might prefer to
2523 @dfn{disable} it. This makes the breakpoint inoperative as if it had
2524 been deleted, but remembers the information on the breakpoint so that
2525 you can @dfn{enable} it again later.
2527 You disable and enable breakpoints and watchpoints with the
2528 @code{enable} and @code{disable} commands, optionally specifying one or
2529 more breakpoint numbers as arguments. Use @code{info break} or
2530 @code{info watch} to print a list of breakpoints or watchpoints if you
2531 do not know which numbers to use.
2533 A breakpoint or watchpoint can have any of four different states of
2538 Enabled. The breakpoint stops your program. A breakpoint set
2539 with the @code{break} command starts out in this state.
2541 Disabled. The breakpoint has no effect on your program.
2543 Enabled once. The breakpoint stops your program, but then becomes
2544 disabled. A breakpoint set with the @code{tbreak} command starts out in
2547 Enabled for deletion. The breakpoint stops your program, but
2548 immediately after it does so it is deleted permanently.
2551 You can use the following commands to enable or disable breakpoints and
2555 @kindex disable breakpoints
2558 @item disable @r{[}breakpoints@r{]} @r{[}@var{bnums}@dots{}@r{]}
2559 Disable the specified breakpoints---or all breakpoints, if none are
2560 listed. A disabled breakpoint has no effect but is not forgotten. All
2561 options such as ignore-counts, conditions and commands are remembered in
2562 case the breakpoint is enabled again later. You may abbreviate
2563 @code{disable} as @code{dis}.
2565 @kindex enable breakpoints
2567 @item enable @r{[}breakpoints@r{]} @r{[}@var{bnums}@dots{}@r{]}
2568 Enable the specified breakpoints (or all defined breakpoints). They
2569 become effective once again in stopping your program.
2571 @item enable @r{[}breakpoints@r{]} once @var{bnums}@dots{}
2572 Enable the specified breakpoints temporarily. @value{GDBN} disables any
2573 of these breakpoints immediately after stopping your program.
2575 @item enable @r{[}breakpoints@r{]} delete @var{bnums}@dots{}
2576 Enable the specified breakpoints to work once, then die. @value{GDBN}
2577 deletes any of these breakpoints as soon as your program stops there.
2580 Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks,
2581 ,Setting breakpoints}), breakpoints that you set are initially enabled;
2582 subsequently, they become disabled or enabled only when you use one of
2583 the commands above. (The command @code{until} can set and delete a
2584 breakpoint of its own, but it does not change the state of your other
2585 breakpoints; see @ref{Continuing and Stepping, ,Continuing and
2589 @subsection Break conditions
2590 @cindex conditional breakpoints
2591 @cindex breakpoint conditions
2593 @c FIXME what is scope of break condition expr? Context where wanted?
2594 @c in particular for a watchpoint?
2595 The simplest sort of breakpoint breaks every time your program reaches a
2596 specified place. You can also specify a @dfn{condition} for a
2597 breakpoint. A condition is just a Boolean expression in your
2598 programming language (@pxref{Expressions, ,Expressions}). A breakpoint with
2599 a condition evaluates the expression each time your program reaches it,
2600 and your program stops only if the condition is @emph{true}.
2602 This is the converse of using assertions for program validation; in that
2603 situation, you want to stop when the assertion is violated---that is,
2604 when the condition is false. In C, if you want to test an assertion expressed
2605 by the condition @var{assert}, you should set the condition
2606 @samp{! @var{assert}} on the appropriate breakpoint.
2608 Conditions are also accepted for watchpoints; you may not need them,
2609 since a watchpoint is inspecting the value of an expression anyhow---but
2610 it might be simpler, say, to just set a watchpoint on a variable name,
2611 and specify a condition that tests whether the new value is an interesting
2614 Break conditions can have side effects, and may even call functions in
2615 your program. This can be useful, for example, to activate functions
2616 that log program progress, or to use your own print functions to
2617 format special data structures. The effects are completely predictable
2618 unless there is another enabled breakpoint at the same address. (In
2619 that case, @value{GDBN} might see the other breakpoint first and stop your
2620 program without checking the condition of this one.) Note that
2621 breakpoint commands are usually more convenient and flexible for the
2622 purpose of performing side effects when a breakpoint is reached
2623 (@pxref{Break Commands, ,Breakpoint command lists}).
2625 Break conditions can be specified when a breakpoint is set, by using
2626 @samp{if} in the arguments to the @code{break} command. @xref{Set
2627 Breaks, ,Setting breakpoints}. They can also be changed at any time
2628 with the @code{condition} command. The @code{watch} command does not
2629 recognize the @code{if} keyword; @code{condition} is the only way to
2630 impose a further condition on a watchpoint.
2634 @item condition @var{bnum} @var{expression}
2635 Specify @var{expression} as the break condition for breakpoint or
2636 watchpoint number @var{bnum}. After you set a condition, breakpoint
2637 @var{bnum} stops your program only if the value of @var{expression} is
2638 true (nonzero, in C). When you use @code{condition}, @value{GDBN}
2639 checks @var{expression} immediately for syntactic correctness, and to
2640 determine whether symbols in it have referents in the context of your
2642 @c FIXME so what does GDB do if there is no referent? Moreover, what
2643 @c about watchpoints?
2645 not actually evaluate @var{expression} at the time the @code{condition}
2646 command is given, however. @xref{Expressions, ,Expressions}.
2648 @item condition @var{bnum}
2649 Remove the condition from breakpoint number @var{bnum}. It becomes
2650 an ordinary unconditional breakpoint.
2653 @cindex ignore count (of breakpoint)
2654 A special case of a breakpoint condition is to stop only when the
2655 breakpoint has been reached a certain number of times. This is so
2656 useful that there is a special way to do it, using the @dfn{ignore
2657 count} of the breakpoint. Every breakpoint has an ignore count, which
2658 is an integer. Most of the time, the ignore count is zero, and
2659 therefore has no effect. But if your program reaches a breakpoint whose
2660 ignore count is positive, then instead of stopping, it just decrements
2661 the ignore count by one and continues. As a result, if the ignore count
2662 value is @var{n}, the breakpoint does not stop the next @var{n} times
2663 your program reaches it.
2667 @item ignore @var{bnum} @var{count}
2668 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
2669 The next @var{count} times the breakpoint is reached, your program's
2670 execution does not stop; other than to decrement the ignore count, @value{GDBN}
2673 To make the breakpoint stop the next time it is reached, specify
2676 When you use @code{continue} to resume execution of your program from a
2677 breakpoint, you can specify an ignore count directly as an argument to
2678 @code{continue}, rather than using @code{ignore}. @xref{Continuing and
2679 Stepping,,Continuing and stepping}.
2681 If a breakpoint has a positive ignore count and a condition, the
2682 condition is not checked. Once the ignore count reaches zero,
2683 @value{GDBN} resumes checking the condition.
2685 You could achieve the effect of the ignore count with a condition such
2686 as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that
2687 is decremented each time. @xref{Convenience Vars, ,Convenience
2691 @node Break Commands
2692 @subsection Breakpoint command lists
2694 @cindex breakpoint commands
2695 You can give any breakpoint (or watchpoint) a series of commands to
2696 execute when your program stops due to that breakpoint. For example, you
2697 might want to print the values of certain expressions, or enable other
2703 @item commands @r{[}@var{bnum}@r{]}
2704 @itemx @dots{} @var{command-list} @dots{}
2706 Specify a list of commands for breakpoint number @var{bnum}. The commands
2707 themselves appear on the following lines. Type a line containing just
2708 @code{end} to terminate the commands.
2710 To remove all commands from a breakpoint, type @code{commands} and
2711 follow it immediately with @code{end}; that is, give no commands.
2713 With no @var{bnum} argument, @code{commands} refers to the last
2714 breakpoint or watchpoint set (not to the breakpoint most recently
2718 Pressing @key{RET} as a means of repeating the last @value{GDBN} command is
2719 disabled within a @var{command-list}.
2721 You can use breakpoint commands to start your program up again. Simply
2722 use the @code{continue} command, or @code{step}, or any other command
2723 that resumes execution.
2725 Any other commands in the command list, after a command that resumes
2726 execution, are ignored. This is because any time you resume execution
2727 (even with a simple @code{next} or @code{step}), you may encounter
2728 another breakpoint---which could have its own command list, leading to
2729 ambiguities about which list to execute.
2732 If the first command you specify in a command list is @code{silent}, the
2733 usual message about stopping at a breakpoint is not printed. This may
2734 be desirable for breakpoints that are to print a specific message and
2735 then continue. If none of the remaining commands print anything, you
2736 see no sign that the breakpoint was reached. @code{silent} is
2737 meaningful only at the beginning of a breakpoint command list.
2739 The commands @code{echo}, @code{output}, and @code{printf} allow you to
2740 print precisely controlled output, and are often useful in silent
2741 breakpoints. @xref{Output, ,Commands for controlled output}.
2743 For example, here is how you could use breakpoint commands to print the
2744 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
2750 printf "x is %d\n",x
2755 One application for breakpoint commands is to compensate for one bug so
2756 you can test for another. Put a breakpoint just after the erroneous line
2757 of code, give it a condition to detect the case in which something
2758 erroneous has been done, and give it commands to assign correct values
2759 to any variables that need them. End with the @code{continue} command
2760 so that your program does not stop, and start with the @code{silent}
2761 command so that no output is produced. Here is an example:
2773 @node Breakpoint Menus
2774 @subsection Breakpoint menus
2776 @cindex symbol overloading
2778 Some programming languages (notably C++) permit a single function name
2779 to be defined several times, for application in different contexts.
2780 This is called @dfn{overloading}. When a function name is overloaded,
2781 @samp{break @var{function}} is not enough to tell @value{GDBN} where you want
2782 a breakpoint. If you realize this is a problem, you can use
2783 something like @samp{break @var{function}(@var{types})} to specify which
2784 particular version of the function you want. Otherwise, @value{GDBN} offers
2785 you a menu of numbered choices for different possible breakpoints, and
2786 waits for your selection with the prompt @samp{>}. The first two
2787 options are always @samp{[0] cancel} and @samp{[1] all}. Typing @kbd{1}
2788 sets a breakpoint at each definition of @var{function}, and typing
2789 @kbd{0} aborts the @code{break} command without setting any new
2792 For example, the following session excerpt shows an attempt to set a
2793 breakpoint at the overloaded symbol @code{String::after}.
2794 We choose three particular definitions of that function name:
2796 @c FIXME! This is likely to change to show arg type lists, at least
2798 (@value{GDBP}) b String::after
2801 [2] file:String.cc; line number:867
2802 [3] file:String.cc; line number:860
2803 [4] file:String.cc; line number:875
2804 [5] file:String.cc; line number:853
2805 [6] file:String.cc; line number:846
2806 [7] file:String.cc; line number:735
2808 Breakpoint 1 at 0xb26c: file String.cc, line 867.
2809 Breakpoint 2 at 0xb344: file String.cc, line 875.
2810 Breakpoint 3 at 0xafcc: file String.cc, line 846.
2811 Multiple breakpoints were set.
2812 Use the "delete" command to delete unwanted
2818 @c @ifclear BARETARGET
2819 @c @node Error in Breakpoints
2820 @c @subsection ``Cannot insert breakpoints''
2822 @c FIXME!! 14/6/95 Is there a real example of this? Let's use it.
2824 @c Under some operating systems, breakpoints cannot be used in a program if
2825 @c any other process is running that program. In this situation,
2826 @c attempting to run or continue a program with a breakpoint causes
2827 @c @value{GDBN} to stop the other process.
2829 @c When this happens, you have three ways to proceed:
2833 @c Remove or disable the breakpoints, then continue.
2836 @c Suspend @value{GDBN}, and copy the file containing your program to a new
2837 @c name. Resume @value{GDBN} and use the @code{exec-file} command to specify
2838 @c that @value{GDBN} should run your program under that name.
2839 @c Then start your program again.
2842 @c Relink your program so that the text segment is nonsharable, using the
2843 @c linker option @samp{-N}. The operating system limitation may not apply
2844 @c to nonsharable executables.
2848 @node Continuing and Stepping
2849 @section Continuing and stepping
2853 @cindex resuming execution
2854 @dfn{Continuing} means resuming program execution until your program
2855 completes normally. In contrast, @dfn{stepping} means executing just
2856 one more ``step'' of your program, where ``step'' may mean either one
2857 line of source code, or one machine instruction (depending on what
2858 particular command you use). Either when continuing
2859 or when stepping, your program may stop even sooner, due to
2864 a breakpoint or a signal. (If due to a signal, you may want to use
2865 @code{handle}, or use @samp{signal 0} to resume execution.
2866 @xref{Signals, ,Signals}.)
2873 @item continue @r{[}@var{ignore-count}@r{]}
2874 @itemx c @r{[}@var{ignore-count}@r{]}
2875 @itemx fg @r{[}@var{ignore-count}@r{]}
2876 Resume program execution, at the address where your program last stopped;
2877 any breakpoints set at that address are bypassed. The optional argument
2878 @var{ignore-count} allows you to specify a further number of times to
2879 ignore a breakpoint at this location; its effect is like that of
2880 @code{ignore} (@pxref{Conditions, ,Break conditions}).
2882 The argument @var{ignore-count} is meaningful only when your program
2883 stopped due to a breakpoint. At other times, the argument to
2884 @code{continue} is ignored.
2886 The synonyms @code{c} and @code{fg} are provided purely for convenience,
2887 and have exactly the same behavior as @code{continue}.
2890 To resume execution at a different place, you can use @code{return}
2891 (@pxref{Returning, ,Returning from a function}) to go back to the
2892 calling function; or @code{jump} (@pxref{Jumping, ,Continuing at a
2893 different address}) to go to an arbitrary location in your program.
2895 A typical technique for using stepping is to set a breakpoint
2897 (@pxref{Breakpoints, ,Breakpoints; watchpoints; and exceptions})
2900 (@pxref{Breakpoints, ,Breakpoints and watchpoints})
2903 beginning of the function or the section of your program where a
2904 problem is believed to lie, run your program until it stops at that
2905 breakpoint, and then step through the suspect area, examining the
2906 variables that are interesting, until you see the problem happen.
2912 Continue running your program until control reaches a different source
2913 line, then stop it and return control to @value{GDBN}. This command is
2914 abbreviated @code{s}.
2917 @c "without debugging information" is imprecise; actually "without line
2918 @c numbers in the debugging information". (gcc -g1 has debugging info but
2919 @c not line numbers). But it seems complex to try to make that
2920 @c distinction here.
2921 @emph{Warning:} If you use the @code{step} command while control is
2922 within a function that was compiled without debugging information,
2923 execution proceeds until control reaches a function that does have
2924 debugging information. Likewise, it will not step into a function which
2925 is compiled without debugging information. To step through functions
2926 without debugging information, use the @code{stepi} command, described
2930 The @code{step} command now only stops at the first instruction of a
2931 source line. This prevents the multiple stops that used to occur in
2932 switch statements, for loops, etc. @code{step} continues to stop if a
2933 function that has debugging information is called within the line.
2935 Also, the @code{step} command now only enters a subroutine if there is line
2936 number information for the subroutine. Otherwise it acts like the
2937 @code{next} command. This avoids problems when using @code{cc -gl}
2938 on MIPS machines. Previously, @code{step} entered subroutines if there
2939 was any debugging information about the routine.
2941 @item step @var{count}
2942 Continue running as in @code{step}, but do so @var{count} times. If a
2943 breakpoint is reached,
2945 or a signal not related to stepping occurs before @var{count} steps,
2947 stepping stops right away.
2951 @item next @r{[}@var{count}@r{]}
2952 Continue to the next source line in the current (innermost) stack frame.
2953 This is similar to @code{step}, but function calls that appear within the line
2954 of code are executed without stopping. Execution stops when control
2955 reaches a different line of code at the original stack level that was
2956 executing when you gave the @code{next} command. This command is abbreviated
2959 An argument @var{count} is a repeat count, as for @code{step}.
2962 @c FIX ME!! Do we delete this, or is there a way it fits in with
2963 @c the following paragraph? --- Vctoria
2965 @c @code{next} within a function that lacks debugging information acts like
2966 @c @code{step}, but any function calls appearing within the code of the
2967 @c function are executed without stopping.
2969 The @code{next} command now only stops at the first instruction of a
2970 source line. This prevents the multiple stops that used to occur in
2971 swtch statements, for loops, etc.
2975 Continue running until just after function in the selected stack frame
2976 returns. Print the returned value (if any).
2978 Contrast this with the @code{return} command (@pxref{Returning,
2979 ,Returning from a function}).
2985 Continue running until a source line past the current line, in the
2986 current stack frame, is reached. This command is used to avoid single
2987 stepping through a loop more than once. It is like the @code{next}
2988 command, except that when @code{until} encounters a jump, it
2989 automatically continues execution until the program counter is greater
2990 than the address of the jump.
2992 This means that when you reach the end of a loop after single stepping
2993 though it, @code{until} makes your program continue execution until it
2994 exits the loop. In contrast, a @code{next} command at the end of a loop
2995 simply steps back to the beginning of the loop, which forces you to step
2996 through the next iteration.
2998 @code{until} always stops your program if it attempts to exit the current
3001 @code{until} may produce somewhat counterintuitive results if the order
3002 of machine code does not match the order of the source lines. For
3003 example, in the following excerpt from a debugging session, the @code{f}
3004 (@code{frame}) command shows that execution is stopped at line
3005 @code{206}; yet when we use @code{until}, we get to line @code{195}:
3009 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
3011 (@value{GDBP}) until
3012 195 for ( ; argc > 0; NEXTARG) @{
3015 This happened because, for execution efficiency, the compiler had
3016 generated code for the loop closure test at the end, rather than the
3017 start, of the loop---even though the test in a C @code{for}-loop is
3018 written before the body of the loop. The @code{until} command appeared
3019 to step back to the beginning of the loop when it advanced to this
3020 expression; however, it has not really gone to an earlier
3021 statement---not in terms of the actual machine code.
3023 @code{until} with no argument works by means of single
3024 instruction stepping, and hence is slower than @code{until} with an
3027 @item until @var{location}
3028 @itemx u @var{location}
3029 Continue running your program until either the specified location is
3030 reached, or the current stack frame returns. @var{location} is any of
3031 the forms of argument acceptable to @code{break} (@pxref{Set Breaks,
3032 ,Setting breakpoints}). This form of the command uses breakpoints,
3033 and hence is quicker than @code{until} without an argument.
3039 Execute one machine instruction, then stop and return to the debugger.
3041 It is often useful to do @samp{display/i $pc} when stepping by machine
3042 instructions. This makes @value{GDBN} automatically display the next
3043 instruction to be executed, each time your program stops. @xref{Auto
3044 Display,, Automatic display}.
3046 An argument is a repeat count, as in @code{step}.
3053 Execute one machine instruction, but if it is a function call,
3054 proceed until the function returns.
3056 An argument is a repeat count, as in @code{next}.
3064 A signal is an asynchronous event that can happen in a program. The
3065 operating system defines the possible kinds of signals, and gives each
3066 kind a name and a number. For example, in Unix @code{SIGINT} is the
3067 signal a program gets when you type an interrupt (often @kbd{C-c});
3068 @code{SIGSEGV} is the signal a program gets from referencing a place in
3069 memory far away from all the areas in use; @code{SIGALRM} occurs when
3070 the alarm clock timer goes off (which happens only if your program has
3071 requested an alarm).
3073 @cindex fatal signals
3074 Some signals, including @code{SIGALRM}, are a normal part of the
3075 functioning of your program. Others, such as @code{SIGSEGV}, indicate
3076 errors; these signals are @dfn{fatal} (kill your program immediately) if the
3077 program has not specified in advance some other way to handle the signal.
3078 @code{SIGINT} does not indicate an error in your program, but it is normally
3079 fatal so it can carry out the purpose of the interrupt: to kill the program.
3081 @value{GDBN} has the ability to detect any occurrence of a signal in your
3082 program. You can tell @value{GDBN} in advance what to do for each kind of
3085 @cindex handling signals
3086 Normally, @value{GDBN} is set up to ignore non-erroneous signals like @code{SIGALRM}
3087 (so as not to interfere with their role in the functioning of your program)
3088 but to stop your program immediately whenever an error signal happens.
3089 You can change these settings with the @code{handle} command.
3092 @kindex info signals
3094 Print a table of all the kinds of signals and how @value{GDBN} has been told to
3095 handle each one. You can use this to see the signal numbers of all
3096 the defined types of signals.
3098 @code{info handle} is the new alias for @code{info signals}.
3101 @item handle @var{signal} @var{keywords}@dots{}
3102 Change the way @value{GDBN} handles signal @var{signal}. @var{signal} can
3103 be the number of a signal or its name (with or without the @samp{SIG} at the
3104 beginning). The @var{keywords} say what change to make.
3108 The keywords allowed by the @code{handle} command can be abbreviated.
3109 Their full names are:
3113 @value{GDBN} should not stop your program when this signal happens. It may
3114 still print a message telling you that the signal has come in.
3117 @value{GDBN} should stop your program when this signal happens. This implies
3118 the @code{print} keyword as well.
3121 @value{GDBN} should print a message when this signal happens.
3124 @value{GDBN} should not mention the occurrence of the signal at all. This
3125 implies the @code{nostop} keyword as well.
3128 @value{GDBN} should allow your program to see this signal; your program
3129 can handle the signal, or else it may terminate if the signal is fatal
3133 @value{GDBN} should not allow your program to see this signal.
3137 When a signal stops your program, the signal is not visible until you
3138 continue. Your program sees the signal then, if @code{pass} is in
3139 effect for the signal in question @emph{at that time}. In other words,
3140 after @value{GDBN} reports a signal, you can use the @code{handle}
3141 command with @code{pass} or @code{nopass} to control whether your
3142 program sees that signal when you continue.
3144 You can also use the @code{signal} command to prevent your program from
3145 seeing a signal, or cause it to see a signal it normally would not see,
3146 or to give it any signal at any time. For example, if your program stopped
3147 due to some sort of memory reference error, you might store correct
3148 values into the erroneous variables and continue, hoping to see more
3149 execution; but your program would probably terminate immediately as
3150 a result of the fatal signal once it saw the signal. To prevent this,
3151 you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your
3157 @section Stopping and starting multi-thread programs
3159 When your program has multiple threads (@pxref{Threads,, Debugging
3160 programs with multiple threads}), you can choose whether to set
3161 breakpoints on all threads, or on a particular thread.
3164 @cindex breakpoints and threads
3165 @cindex thread breakpoints
3166 @kindex break @dots{} thread @var{threadno}
3167 @item break @var{linespec} thread @var{threadno}
3168 @itemx break @var{linespec} thread @var{threadno} if @dots{}
3169 @var{linespec} specifies source lines; there are several ways of
3170 writing them, but the effect is always to specify some source line.
3172 Use the qualifier @samp{thread @var{threadno}} with a breakpoint command
3173 to specify that you only want @value{GDBN} to stop the program when a
3174 particular thread reaches this breakpoint. @var{threadno} is one of the
3175 numeric thread identifiers assigned by @value{GDBN}, shown in the first
3176 column of the @samp{info threads} display.
3178 If you do not specify @samp{thread @var{threadno}} when you set a
3179 breakpoint, the breakpoint applies to @emph{all} threads of your
3182 You can use the @code{thread} qualifier on conditional breakpoints as
3183 well; in this case, place @samp{thread @var{threadno}} before the
3184 breakpoint condition, like this:
3187 (gdb) break frik.c:13 thread 28 if bartab > lim
3192 @cindex stopped threads
3193 @cindex threads, stopped
3194 Whenever your program stops under @value{GDBN} for any reason,
3195 @emph{all} threads of execution stop, not just the current thread. This
3196 allows you to examine the overall state of the program, including
3197 switching between threads, without worrying that things may change
3200 @cindex continuing threads
3201 @cindex threads, continuing
3202 Conversely, whenever you restart the program, @emph{all} threads start
3203 executing. @emph{This is true even when single-stepping} with commands
3204 like @code{step} or @code{next}.
3206 In particular, @value{GDBN} cannot single-step all threads in lockstep.
3207 Since thread scheduling is up to your debugging target's operating
3208 system (not controlled by @value{GDBN}), other threads may
3209 execute more than one statement while the current thread completes a
3210 single step. Moreover, in general other threads stop in the middle of a
3211 statement, rather than at a clean statement boundary, when the program
3214 You might even find your program stopped in another thread after
3215 continuing or even single-stepping. This happens whenever some other
3216 thread runs into a breakpoint, a signal, or an exception before the
3217 first thread completes whatever you requested.
3221 @chapter Examining the Stack
3223 When your program has stopped, the first thing you need to know is where it
3224 stopped and how it got there.
3227 Each time your program performs a function call, information about the call
3229 That information includes the location of the call in your program,
3230 the arguments of the call,
3231 and the local variables of the function being called.
3232 The information is saved in a block of data called a @dfn{stack frame}.
3233 The stack frames are allocated in a region of memory called the @dfn{call
3236 When your program stops, the @value{GDBN} commands for examining the
3237 stack allow you to see all of this information.
3239 @cindex selected frame
3240 One of the stack frames is @dfn{selected} by @value{GDBN} and many
3241 @value{GDBN} commands refer implicitly to the selected frame. In
3242 particular, whenever you ask @value{GDBN} for the value of a variable in
3243 your program, the value is found in the selected frame. There are
3244 special @value{GDBN} commands to select whichever frame you are
3245 interested in. @xref{Selection, ,Selecting a frame}.
3247 When your program stops, @value{GDBN} automatically selects the
3248 currently executing frame and describes it briefly, similar to the
3249 @code{frame} command (@pxref{Frame Info, ,Information about a frame}).
3252 * Frames:: Stack frames
3253 * Backtrace:: Backtraces
3254 * Selection:: Selecting a frame
3255 * Frame Info:: Information on a frame
3257 * MIPS Stack:: MIPS machines and the function stack
3262 @section Stack frames
3266 The call stack is divided up into contiguous pieces called @dfn{stack
3267 frames}, or @dfn{frames} for short; each frame is the data associated
3268 with one call to one function. The frame contains the arguments given
3269 to the function, the function's local variables, and the address at
3270 which the function is executing.
3272 @cindex initial frame
3273 @cindex outermost frame
3274 @cindex innermost frame
3275 When your program is started, the stack has only one frame, that of the
3276 function @code{main}. This is called the @dfn{initial} frame or the
3277 @dfn{outermost} frame. Each time a function is called, a new frame is
3278 made. Each time a function returns, the frame for that function invocation
3279 is eliminated. If a function is recursive, there can be many frames for
3280 the same function. The frame for the function in which execution is
3281 actually occurring is called the @dfn{innermost} frame. This is the most
3282 recently created of all the stack frames that still exist.
3284 @cindex frame pointer
3285 Inside your program, stack frames are identified by their addresses. A
3286 stack frame consists of many bytes, each of which has its own address; each
3287 kind of computer has a convention for choosing one byte whose
3288 address serves as the address of the frame. Usually this address is kept
3289 in a register called the @dfn{frame pointer register} while execution is
3290 going on in that frame.
3292 @cindex frame number
3293 @value{GDBN} assigns numbers to all existing stack frames, starting with
3294 zero for the innermost frame, one for the frame that called it,
3295 and so on upward. These numbers do not really exist in your program;
3296 they are assigned by @value{GDBN} to give you a way of designating stack
3297 frames in @value{GDBN} commands.
3299 @c below produces an acceptable overful hbox. --mew 13aug1993
3300 @cindex frameless execution
3301 Some compilers provide a way to compile functions so that they operate
3302 without stack frames. (For example, the @code{@value{GCC}} option
3303 @samp{-fomit-frame-pointer} generates functions without a frame.)
3304 This is occasionally done with heavily used library functions to save
3305 the frame setup time. @value{GDBN} has limited facilities for dealing
3306 with these function invocations. If the innermost function invocation
3307 has no stack frame, @value{GDBN} nevertheless regards it as though
3308 it had a separate frame, which is numbered zero as usual, allowing
3309 correct tracing of the function call chain. However, @value{GDBN} has
3310 no provision for frameless functions elsewhere in the stack.
3314 @item frame @var{args}
3315 The @code{frame} command allows you to move from one stack frame to another,
3316 and to print the stack frame you select. @var{args} may be either the
3317 address of the frame or the stack frame number. Without an argument,
3318 @code{frame} prints the current stack frame.
3320 @kindex select-frame
3322 The @code{select-frame} command allows you to move from one stack frame
3323 to another without printing the frame. This is the silent version of
3330 A backtrace is a summary of how your program got where it is. It shows one
3331 line per frame, for many frames, starting with the currently executing
3332 frame (frame zero), followed by its caller (frame one), and on up the
3340 Print a backtrace of the entire stack: one line per frame for all
3341 frames in the stack.
3343 You can stop the backtrace at any time by typing the system interrupt
3344 character, normally @kbd{C-c}.
3346 @item backtrace @var{n}
3348 Similar, but print only the innermost @var{n} frames.
3350 @item backtrace -@var{n}
3352 Similar, but print only the outermost @var{n} frames.
3358 The names @code{where} and @code{info stack} (abbreviated @code{info s})
3359 are additional aliases for @code{backtrace}.
3361 Each line in the backtrace shows the frame number and the function name.
3362 The program counter value is also shown---unless you use @code{set
3363 print address off}. The backtrace also shows the source file name and
3364 line number, as well as the arguments to the function. The program
3365 counter value is omitted if it is at the beginning of the code for that
3368 Here is an example of a backtrace. It was made with the command
3369 @samp{bt 3}, so it shows the innermost three frames.
3373 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
3375 #1 0x6e38 in expand_macro (sym=0x2b600) at macro.c:242
3376 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
3378 (More stack frames follow...)
3383 The display for frame zero does not begin with a program counter
3384 value, indicating that your program has stopped at the beginning of the
3385 code for line @code{993} of @code{builtin.c}.
3388 @section Selecting a frame
3390 Most commands for examining the stack and other data in your program work on
3391 whichever stack frame is selected at the moment. Here are the commands for
3392 selecting a stack frame; all of them finish by printing a brief description
3393 of the stack frame just selected.
3400 Select frame number @var{n}. Recall that frame zero is the innermost
3401 (currently executing) frame, frame one is the frame that called the
3402 innermost one, and so on. The highest-numbered frame is the one for
3405 @item frame @var{addr}
3407 Select the frame at address @var{addr}. This is useful mainly if the
3408 chaining of stack frames has been damaged by a bug, making it
3409 impossible for @value{GDBN} to assign numbers properly to all frames. In
3410 addition, this can be useful when your program has multiple stacks and
3411 switches between them.
3413 @ifclear H8EXCLUSIVE
3414 On the SPARC architecture, @code{frame} needs two addresses to
3415 select an arbitrary frame: a frame pointer and a stack pointer.
3417 On the MIPS and Alpha architecture, it needs two addresses: a stack
3418 pointer and a program counter.
3420 On the 29k architecture, it needs three addresses: a register stack
3421 pointer, a program counter, and a memory stack pointer.
3422 @c note to future updaters: this is conditioned on a flag
3423 @c SETUP_ARBITRARY_FRAME in the tm-*.h files. The above is up to date
3424 @c as of 27 Jan 1994.
3429 Move @var{n} frames up the stack. For positive numbers @var{n}, this
3430 advances toward the outermost frame, to higher frame numbers, to frames
3431 that have existed longer. @var{n} defaults to one.
3436 Move @var{n} frames down the stack. For positive numbers @var{n}, this
3437 advances toward the innermost frame, to lower frame numbers, to frames
3438 that were created more recently. @var{n} defaults to one. You may
3439 abbreviate @code{down} as @code{do}.
3442 All of these commands end by printing two lines of output describing the
3443 frame. The first line shows the frame number, the function name, the
3444 arguments, and the source file and line number of execution in that
3445 frame. The second line shows the text of that source line.
3453 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc)
3455 10 read_input_file (argv[i]);
3459 After such a printout, the @code{list} command with no arguments
3460 prints ten lines centered on the point of execution in the frame.
3461 @xref{List, ,Printing source lines}.
3464 @kindex down-silently
3466 @item up-silently @var{n}
3467 @itemx down-silently @var{n}
3468 These two commands are variants of @code{up} and @code{down},
3469 respectively; they differ in that they do their work silently, without
3470 causing display of the new frame. They are intended primarily for use
3471 in @value{GDBN} command scripts, where the output might be unnecessary and
3476 @section Information about a frame
3478 There are several other commands to print information about the selected
3484 When used without any argument, this command does not change which
3485 frame is selected, but prints a brief description of the currently
3486 selected stack frame. It can be abbreviated @code{f}. With an
3487 argument, this command is used to select a stack frame.
3488 @xref{Selection, ,Selecting a frame}.
3494 This command prints a verbose description of the selected stack frame,
3499 the address of the frame
3501 the address of the next frame down (called by this frame)
3503 the address of the next frame up (caller of this frame)
3505 the language in which the source code corresponding to this frame is written
3507 the address of the frame's arguments
3509 the program counter saved in it (the address of execution in the caller frame)
3511 which registers were saved in the frame
3514 @noindent The verbose description is useful when
3515 something has gone wrong that has made the stack format fail to fit
3516 the usual conventions.
3518 @item info frame @var{addr}
3519 @itemx info f @var{addr}
3520 Print a verbose description of the frame at address @var{addr}, without
3521 selecting that frame. The selected frame remains unchanged by this
3522 command. This requires the same kind of address (more than one for some
3523 architectures) that you specify in the @code{frame} command.
3524 @xref{Selection, ,Selecting a frame}.
3528 Print the arguments of the selected frame, each on a separate line.
3532 Print the local variables of the selected frame, each on a separate
3533 line. These are all variables (declared either static or automatic)
3534 accessible at the point of execution of the selected frame.
3538 @cindex catch exceptions
3539 @cindex exception handlers
3541 Print a list of all the exception handlers that are active in the
3542 current stack frame at the current point of execution. To see other
3543 exception handlers, visit the associated frame (using the @code{up},
3544 @code{down}, or @code{frame} commands); then type @code{info catch}.
3545 @xref{Exception Handling, ,Breakpoints and exceptions}.
3551 @section MIPS machines and the function stack
3553 @cindex stack on MIPS
3555 MIPS based computers use an unusual stack frame, which sometimes
3556 requires @value{GDBN} to search backward in the object code to find the
3557 beginning of a function.
3559 @cindex response time, MIPS debugging
3560 To improve response time (especially for embedded applications, where
3561 @value{GDBN} may be restricted to a slow serial line for this search)
3562 you may want to limit the size of this search, using one of these
3566 @cindex @code{heuristic-fence-post} (MIPS)
3567 @item set heuristic-fence-post @var{limit}
3568 Restrict @value{GDBN} to examining at most @var{limit} bytes in its search
3569 for the beginning of a function. A value of @var{0} (the default)
3570 means there is no limit. However, except for @var{0}, the larger the
3571 limit the more bytes @code{heuristic-fence-post} must search and
3572 therefore the longer it takes to run.
3574 @item show heuristic-fence-post
3575 Display the current limit.
3579 These commands are available @emph{only} when @value{GDBN} is configured
3580 for debugging programs on MIPS processors.
3584 @chapter Examining Source Files
3586 @value{GDBN} can print parts of your program's source, since the debugging
3587 information recorded in the program tells @value{GDBN} what source files were
3588 used to build it. When your program stops, @value{GDBN} spontaneously prints
3589 the line where it stopped. Likewise, when you select a stack frame
3590 (@pxref{Selection, ,Selecting a frame}), @value{GDBN} prints the line where
3591 execution in that frame has stopped. You can print other portions of
3592 source files by explicit command.
3595 If you use @value{GDBN} through its @sc{gnu} Emacs interface, you may prefer
3597 Emacs facilities to view source; @pxref{Emacs, ,Using @value{GDBN} under @sc{gnu} Emacs}.
3601 * List:: Printing source lines
3603 * Search:: Searching source files
3606 * Source Path:: Specifying source directories
3607 * Machine Code:: Source and machine code
3611 @section Printing source lines
3615 To print lines from a source file, use the @code{list} command
3616 (abbreviated @code{l}). By default, ten lines are printed.
3617 There are several ways to specify what part of the file you want to print.
3619 Here are the forms of the @code{list} command most commonly used:
3622 @item list @var{linenum}
3623 Print lines centered around line number @var{linenum} in the
3624 current source file.
3626 @item list @var{function}
3627 Print lines centered around the beginning of function
3631 Print more lines. If the last lines printed were printed with a
3632 @code{list} command, this prints lines following the last lines
3633 printed; however, if the last line printed was a solitary line printed
3634 as part of displaying a stack frame (@pxref{Stack, ,Examining the
3635 Stack}), this prints lines centered around that line.
3638 Print lines just before the lines last printed.
3641 By default, @value{GDBN} prints ten source lines with any of these forms of
3642 the @code{list} command. You can change this using @code{set listsize}:
3645 @kindex set listsize
3646 @item set listsize @var{count}
3647 Make the @code{list} command display @var{count} source lines (unless
3648 the @code{list} argument explicitly specifies some other number).
3650 @kindex show listsize
3652 Display the number of lines that @code{list} prints.
3655 Repeating a @code{list} command with @key{RET} discards the argument,
3656 so it is equivalent to typing just @code{list}. This is more useful
3657 than listing the same lines again. An exception is made for an
3658 argument of @samp{-}; that argument is preserved in repetition so that
3659 each repetition moves up in the source file.
3662 In general, the @code{list} command expects you to supply zero, one or two
3663 @dfn{linespecs}. Linespecs specify source lines; there are several ways
3664 of writing them but the effect is always to specify some source line.
3665 Here is a complete description of the possible arguments for @code{list}:
3668 @item list @var{linespec}
3669 Print lines centered around the line specified by @var{linespec}.
3671 @item list @var{first},@var{last}
3672 Print lines from @var{first} to @var{last}. Both arguments are
3675 @item list ,@var{last}
3676 Print lines ending with @var{last}.
3678 @item list @var{first},
3679 Print lines starting with @var{first}.
3682 Print lines just after the lines last printed.
3685 Print lines just before the lines last printed.
3688 As described in the preceding table.
3691 Here are the ways of specifying a single source line---all the
3696 Specifies line @var{number} of the current source file.
3697 When a @code{list} command has two linespecs, this refers to
3698 the same source file as the first linespec.
3701 Specifies the line @var{offset} lines after the last line printed.
3702 When used as the second linespec in a @code{list} command that has
3703 two, this specifies the line @var{offset} lines down from the
3707 Specifies the line @var{offset} lines before the last line printed.
3709 @item @var{filename}:@var{number}
3710 Specifies line @var{number} in the source file @var{filename}.
3712 @item @var{function}
3713 Specifies the line that begins the body of the function @var{function}.
3714 For example: in C, this is the line with the open brace.
3716 @item @var{filename}:@var{function}
3717 Specifies the line of the open-brace that begins the body of the
3718 function @var{function} in the file @var{filename}. You only need the
3719 file name with a function name to avoid ambiguity when there are
3720 identically named functions in different source files.
3722 @item *@var{address}
3723 Specifies the line containing the program address @var{address}.
3724 @var{address} may be any expression.
3729 @section Searching source files
3731 @kindex reverse-search
3733 There are two commands for searching through the current source file for a
3738 @kindex forward-search
3739 @item forward-search @var{regexp}
3740 @itemx search @var{regexp}
3741 The command @samp{forward-search @var{regexp}} checks each line,
3742 starting with the one following the last line listed, for a match for
3743 @var{regexp}. It lists the line that is found. You can use the
3744 synonym @samp{search @var{regexp}} or abbreviate the command name as
3747 @item reverse-search @var{regexp}
3748 The command @samp{reverse-search @var{regexp}} checks each line, starting
3749 with the one before the last line listed and going backward, for a match
3750 for @var{regexp}. It lists the line that is found. You can abbreviate
3751 this command as @code{rev}.
3756 @section Specifying source directories
3759 @cindex directories for source files
3760 Executable programs sometimes do not record the directories of the source
3761 files from which they were compiled, just the names. Even when they do,
3762 the directories could be moved between the compilation and your debugging
3763 session. @value{GDBN} has a list of directories to search for source files;
3764 this is called the @dfn{source path}. Each time @value{GDBN} wants a source file,
3765 it tries all the directories in the list, in the order they are present
3766 in the list, until it finds a file with the desired name. Note that
3767 the executable search path is @emph{not} used for this purpose. Neither is
3768 the current working directory, unless it happens to be in the source
3771 If @value{GDBN} cannot find a source file in the source path, and the
3772 object program records a directory, @value{GDBN} tries that directory
3773 too. If the source path is empty, and there is no record of the
3774 compilation directory, @value{GDBN} looks in the current directory as a
3777 Whenever you reset or rearrange the source path, @value{GDBN} clears out
3778 any information it has cached about where source files are found and where
3779 each line is in the file.
3783 When you start @value{GDBN}, its source path is empty.
3784 To add other directories, use the @code{directory} command.
3787 @item directory @var{dirname} @dots{}
3788 @item dir @var{dirname} @dots{}
3789 Add directory @var{dirname} to the front of the source path. Several
3790 directory names may be given to this command, separated by @samp{:} or
3791 whitespace. You may specify a directory that is already in the source
3792 path; this moves it forward, so @value{GDBN} searches it sooner.
3798 @cindex compilation directory
3799 @cindex current directory
3800 @cindex working directory
3801 @cindex directory, current
3802 @cindex directory, compilation
3803 You can use the string @samp{$cdir} to refer to the compilation
3804 directory (if one is recorded), and @samp{$cwd} to refer to the current
3805 working directory. @samp{$cwd} is not the same as @samp{.}---the former
3806 tracks the current working directory as it changes during your @value{GDBN}
3807 session, while the latter is immediately expanded to the current
3808 directory at the time you add an entry to the source path.
3811 Reset the source path to empty again. This requires confirmation.
3813 @c RET-repeat for @code{directory} is explicitly disabled, but since
3814 @c repeating it would be a no-op we do not say that. (thanks to RMS)
3816 @item show directories
3817 @kindex show directories
3818 Print the source path: show which directories it contains.
3821 If your source path is cluttered with directories that are no longer of
3822 interest, @value{GDBN} may sometimes cause confusion by finding the wrong
3823 versions of source. You can correct the situation as follows:
3827 Use @code{directory} with no argument to reset the source path to empty.
3830 Use @code{directory} with suitable arguments to reinstall the
3831 directories you want in the source path. You can add all the
3832 directories in one command.
3836 @section Source and machine code
3838 You can use the command @code{info line} to map source lines to program
3839 addresses (and vice versa), and the command @code{disassemble} to display
3840 a range of addresses as machine instructions. When run under @sc{gnu} Emacs
3841 mode, the @code{info line} command now causes the arrow to point to the
3842 line specified. Also, @code{info line} prints addresses in symbolic form as
3847 @item info line @var{linespec}
3848 Print the starting and ending addresses of the compiled code for
3849 source line @var{linespec}. You can specify source lines in any of
3850 the ways understood by the @code{list} command (@pxref{List, ,Printing
3854 For example, we can use @code{info line} to discover the location of
3855 the object code for the first line of function
3856 @code{m4_changequote}:
3859 (@value{GDBP}) info line m4_changecom
3860 Line 895 of "builtin.c" starts at pc 0x634c and ends at 0x6350.
3864 We can also inquire (using @code{*@var{addr}} as the form for
3865 @var{linespec}) what source line covers a particular address:
3867 (@value{GDBP}) info line *0x63ff
3868 Line 926 of "builtin.c" starts at pc 0x63e4 and ends at 0x6404.
3871 @cindex @code{$_} and @code{info line}
3872 After @code{info line}, the default address for the @code{x} command
3873 is changed to the starting address of the line, so that @samp{x/i} is
3874 sufficient to begin examining the machine code (@pxref{Memory,
3875 ,Examining memory}). Also, this address is saved as the value of the
3876 convenience variable @code{$_} (@pxref{Convenience Vars, ,Convenience
3881 @cindex assembly instructions
3882 @cindex instructions, assembly
3883 @cindex machine instructions
3884 @cindex listing machine instructions
3886 This specialized command dumps a range of memory as machine
3887 instructions. The default memory range is the function surrounding the
3888 program counter of the selected frame. A single argument to this
3889 command is a program counter value; @value{GDBN} dumps the function
3890 surrounding this value. Two arguments specify a range of addresses
3891 (first inclusive, second exclusive) to dump.
3894 @ifclear H8EXCLUSIVE
3895 We can use @code{disassemble} to inspect the object code
3896 range shown in the last @code{info line} example (the example
3897 shows SPARC machine instructions):
3901 (@value{GDBP}) disas 0x63e4 0x6404
3902 Dump of assembler code from 0x63e4 to 0x6404:
3903 0x63e4 <builtin_init+5340>: ble 0x63f8 <builtin_init+5360>
3904 0x63e8 <builtin_init+5344>: sethi %hi(0x4c00), %o0
3905 0x63ec <builtin_init+5348>: ld [%i1+4], %o0
3906 0x63f0 <builtin_init+5352>: b 0x63fc <builtin_init+5364>
3907 0x63f4 <builtin_init+5356>: ld [%o0+4], %o0
3908 0x63f8 <builtin_init+5360>: or %o0, 0x1a4, %o0
3909 0x63fc <builtin_init+5364>: call 0x9288 <path_search>
3910 0x6400 <builtin_init+5368>: nop
3911 End of assembler dump.
3916 For example, here is the beginning of the output for the
3917 disassembly of a function @code{fact}:
3921 (@value{GDBP}) disas fact
3922 Dump of assembler code for function fact:
3924 0x802c <fact>: 6d f2 mov.w r2,@@-r7
3925 0x802e <fact+2>: 6d f3 mov.w r3,@@-r7
3926 0x8030 <fact+4>: 6d f6 mov.w r6,@@-r7
3927 0x8032 <fact+6>: 0d 76 mov.w r7,r6
3928 0x8034 <fact+8>: 6f 70 00 08 mov.w @@(0x8,r7),r0
3929 0x8038 <fact+12> 19 11 sub.w r1,r1
3937 @kindex set assembly-language
3938 @cindex assembly instructions
3939 @cindex instructions, assembly
3940 @cindex machine instructions
3941 @cindex listing machine instructions
3942 @item set assembly-language @var{instruction-set}
3943 This command selects the instruction set to use when disassembling the program via the
3944 @code{disassemble} or @code{x/i} commands. It is useful for architectures that
3945 have more than one native instruction set.
3947 Currently it is only defined for the Intel x86 family. You can set @var{instruction-set}
3948 to either @code{i386} or @code{i8086}. The default is @code{i386}.
3953 @chapter Examining Data
3955 @cindex printing data
3956 @cindex examining data
3959 @c "inspect" is not quite a synonym if you are using Epoch, which we do not
3960 @c document because it is nonstandard... Under Epoch it displays in a
3961 @c different window or something like that.
3962 The usual way to examine data in your program is with the @code{print}
3963 command (abbreviated @code{p}), or its synonym @code{inspect}.
3965 It evaluates and prints the value of an expression of the language your
3966 program is written in (@pxref{Languages, ,Using @value{GDBN} with Different
3971 @item print @var{exp}
3972 @itemx print /@var{f} @var{exp}
3973 @var{exp} is an expression (in the source language). By default the
3974 value of @var{exp} is printed in a format appropriate to its data type;
3975 you can choose a different format by specifying @samp{/@var{f}}, where
3976 @var{f} is a letter specifying the format; @pxref{Output Formats,,Output
3980 @itemx print /@var{f}
3981 If you omit @var{exp}, @value{GDBN} displays the last value again (from the
3982 @dfn{value history}; @pxref{Value History, ,Value history}). This allows you to
3983 conveniently inspect the same value in an alternative format.
3986 A more low-level way of examining data is with the @code{x} command.
3987 It examines data in memory at a specified address and prints it in a
3988 specified format. @xref{Memory, ,Examining memory}.
3990 If you are interested in information about types, or about how the fields
3995 are declared, use the @code{ptype @var{exp}}
3996 command rather than @code{print}. @xref{Symbols, ,Examining the Symbol Table}.
3999 * Expressions:: Expressions
4000 * Variables:: Program variables
4001 * Arrays:: Artificial arrays
4002 * Output Formats:: Output formats
4003 * Memory:: Examining memory
4004 * Auto Display:: Automatic display
4005 * Print Settings:: Print settings
4006 * Value History:: Value history
4007 * Convenience Vars:: Convenience variables
4008 * Registers:: Registers
4010 * Floating Point Hardware:: Floating point hardware
4015 @section Expressions
4018 @code{print} and many other @value{GDBN} commands accept an expression and
4019 compute its value. Any kind of constant, variable or operator defined
4020 by the programming language you are using is valid in an expression in
4021 @value{GDBN}. This includes conditional expressions, function calls, casts
4022 and string constants. It unfortunately does not include symbols defined
4023 by preprocessor @code{#define} commands.
4025 @value{GDBN} now supports array constants in expressions input by
4026 the user. The syntax is @var{@{element, element@dots{}@}}. For example,
4027 you can now use the command @code{print @{1, 2, 3@}} to build up an array in
4028 memory that is malloc'd in the target program.
4031 Because C is so widespread, most of the expressions shown in examples in
4032 this manual are in C. @xref{Languages, , Using @value{GDBN} with Different
4033 Languages}, for information on how to use expressions in other
4036 In this section, we discuss operators that you can use in @value{GDBN}
4037 expressions regardless of your programming language.
4039 Casts are supported in all languages, not just in C, because it is so
4040 useful to cast a number into a pointer in order to examine a structure
4041 at that address in memory.
4042 @c FIXME: casts supported---Mod2 true?
4045 @value{GDBN} supports these operators, in addition to those common
4046 to programming languages:
4050 @samp{@@} is a binary operator for treating parts of memory as arrays.
4051 @xref{Arrays, ,Artificial arrays}, for more information.
4054 @samp{::} allows you to specify a variable in terms of the file or
4055 function where it is defined. @xref{Variables, ,Program variables}.
4057 @cindex @{@var{type}@}
4058 @cindex type casting memory
4059 @cindex memory, viewing as typed object
4060 @cindex casts, to view memory
4061 @item @{@var{type}@} @var{addr}
4062 Refers to an object of type @var{type} stored at address @var{addr} in
4063 memory. @var{addr} may be any expression whose value is an integer or
4064 pointer (but parentheses are required around binary operators, just as in
4065 a cast). This construct is allowed regardless of what kind of data is
4066 normally supposed to reside at @var{addr}.
4070 @section Program variables
4072 The most common kind of expression to use is the name of a variable
4075 Variables in expressions are understood in the selected stack frame
4076 (@pxref{Selection, ,Selecting a frame}); they must be either:
4087 visible according to the scope rules of the
4088 programming language from the point of execution in that frame
4091 @noindent This means that in the function
4106 you can examine and use the variable @code{a} whenever your program is
4107 executing within the function @code{foo}, but you can only use or
4108 examine the variable @code{b} while your program is executing inside
4109 the block where @code{b} is declared.
4111 @cindex variable name conflict
4112 There is an exception: you can refer to a variable or function whose
4113 scope is a single source file even if the current execution point is not
4114 in this file. But it is possible to have more than one such variable or
4115 function with the same name (in different source files). If that
4116 happens, referring to that name has unpredictable effects. If you wish,
4117 you can specify a static variable in a particular function or file,
4118 using the colon-colon notation:
4122 @c info cannot cope with a :: index entry, but why deprive hard copy readers?
4126 @var{file}::@var{variable}
4127 @var{function}::@var{variable}
4131 Here @var{file} or @var{function} is the name of the context for the
4132 static @var{variable}. In the case of file names, you can use quotes to
4133 make sure @value{GDBN} parses the file name as a single word---for example,
4134 to print a global value of @code{x} defined in @file{f2.c}:
4137 (@value{GDBP}) p 'f2.c'::x
4141 @cindex C++ scope resolution
4142 This use of @samp{::} is very rarely in conflict with the very similar
4143 use of the same notation in C++. @value{GDBN} also supports use of the C++
4144 scope resolution operator in @value{GDBN} expressions.
4145 @c FIXME: Um, so what happens in one of those rare cases where it's in
4149 @cindex wrong values
4150 @cindex variable values, wrong
4152 @emph{Warning:} Occasionally, a local variable may appear to have the
4153 wrong value at certain points in a function---just after entry to a new
4154 scope, and just before exit.
4156 You may see this problem when you are stepping by machine instructions.
4157 This is because, on most machines, it takes more than one instruction to
4158 set up a stack frame (including local variable definitions); if you are
4159 stepping by machine instructions, variables may appear to have the wrong
4160 values until the stack frame is completely built. On exit, it usually
4161 also takes more than one machine instruction to destroy a stack frame;
4162 after you begin stepping through that group of instructions, local
4163 variable definitions may be gone.
4166 @section Artificial arrays
4168 @cindex artificial array
4170 It is often useful to print out several successive objects of the
4171 same type in memory; a section of an array, or an array of
4172 dynamically determined size for which only a pointer exists in the
4175 You can do this by referring to a contiguous span of memory as an
4176 @dfn{artificial array}, using the binary operator @samp{@@}. The left
4177 operand of @samp{@@} should be the first element of the desired array
4178 and be an individual object. The right operand should be the desired length
4179 of the array. The result is an array value whose elements are all of
4180 the type of the left argument. The first element is actually the left
4181 argument; the second element comes from bytes of memory immediately
4182 following those that hold the first element, and so on. Here is an
4183 example. If a program says
4186 int *array = (int *) malloc (len * sizeof (int));
4190 you can print the contents of @code{array} with
4196 The left operand of @samp{@@} must reside in memory. Array values made
4197 with @samp{@@} in this way behave just like other arrays in terms of
4198 subscripting, and are coerced to pointers when used in expressions.
4199 Artificial arrays most often appear in expressions via the value history
4200 (@pxref{Value History, ,Value history}), after printing one out.
4202 Another way to create an artificial array is to use a cast.
4203 This re-interprets a value as if it were an array.
4204 The value need not be in memory:
4206 (@value{GDBP}) p/x (short[2])0x12345678
4207 $1 = @{0x1234, 0x5678@}
4210 As a convenience, if you leave the array length out (as in
4211 @samp{(@var{type})[])@var{value}}) gdb calculates the size to fill
4212 the value (as @samp{sizeof(@var{value})/sizeof(@var{type})}:
4214 (@value{GDBP}) p/x (short[])0x12345678
4215 $2 = @{0x1234, 0x5678@}
4218 Sometimes the artificial array mechanism is not quite enough; in
4219 moderately complex data structures, the elements of interest may not
4220 actually be adjacent---for example, if you are interested in the values
4221 of pointers in an array. One useful work-around in this situation is
4222 to use a convenience variable (@pxref{Convenience Vars, ,Convenience
4223 variables}) as a counter in an expression that prints the first
4224 interesting value, and then repeat that expression via @key{RET}. For
4225 instance, suppose you have an array @code{dtab} of pointers to
4226 structures, and you are interested in the values of a field @code{fv}
4227 in each structure. Here is an example of what you might type:
4237 @node Output Formats
4238 @section Output formats
4240 @cindex formatted output
4241 @cindex output formats
4242 By default, @value{GDBN} prints a value according to its data type. Sometimes
4243 this is not what you want. For example, you might want to print a number
4244 in hex, or a pointer in decimal. Or you might want to view data in memory
4245 at a certain address as a character string or as an instruction. To do
4246 these things, specify an @dfn{output format} when you print a value.
4248 The simplest use of output formats is to say how to print a value
4249 already computed. This is done by starting the arguments of the
4250 @code{print} command with a slash and a format letter. The format
4251 letters supported are:
4255 Regard the bits of the value as an integer, and print the integer in
4259 Print as integer in signed decimal.
4262 Print as integer in unsigned decimal.
4265 Print as integer in octal.
4268 Print as integer in binary. The letter @samp{t} stands for ``two''.
4269 @footnote{@samp{b} cannot be used because these format letters are also
4270 used with the @code{x} command, where @samp{b} stands for ``byte'';
4271 @pxref{Memory,,Examining memory}.}
4274 @cindex unknown address, locating
4275 Print as an address, both absolute in hexadecimal and as an offset from
4276 the nearest preceding symbol. You can use this format used to discover
4277 where (in what function) an unknown address is located:
4280 (@value{GDBP}) p/a 0x54320
4281 $3 = 0x54320 <_initialize_vx+396>
4285 Regard as an integer and print it as a character constant.
4288 Regard the bits of the value as a floating point number and print
4289 using typical floating point syntax.
4292 For example, to print the program counter in hex (@pxref{Registers}), type
4299 Note that no space is required before the slash; this is because command
4300 names in @value{GDBN} cannot contain a slash.
4302 To reprint the last value in the value history with a different format,
4303 you can use the @code{print} command with just a format and no
4304 expression. For example, @samp{p/x} reprints the last value in hex.
4307 @section Examining memory
4309 You can use the command @code{x} (for ``examine'') to examine memory in
4310 any of several formats, independently of your program's data types.
4312 @cindex examining memory
4315 @item x/@var{nfu} @var{addr}
4318 Use the @code{x} command to examine memory.
4321 @var{n}, @var{f}, and @var{u} are all optional parameters that specify how
4322 much memory to display and how to format it; @var{addr} is an
4323 expression giving the address where you want to start displaying memory.
4324 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
4325 Several commands set convenient defaults for @var{addr}.
4328 @item @var{n}, the repeat count
4329 The repeat count is a decimal integer; the default is 1. It specifies
4330 how much memory (counting by units @var{u}) to display.
4331 @c This really is **decimal**; unaffected by 'set radix' as of GDB
4334 @item @var{f}, the display format
4335 The display format is one of the formats used by @code{print},
4336 @samp{s} (null-terminated string), or @samp{i} (machine instruction).
4337 The default is @samp{x} (hexadecimal) initially.
4338 The default changes each time you use either @code{x} or @code{print}.
4340 @item @var{u}, the unit size
4341 The unit size is any of
4347 Halfwords (two bytes).
4349 Words (four bytes). This is the initial default.
4351 Giant words (eight bytes).
4354 Each time you specify a unit size with @code{x}, that size becomes the
4355 default unit the next time you use @code{x}. (For the @samp{s} and
4356 @samp{i} formats, the unit size is ignored and is normally not written.)
4358 @item @var{addr}, starting display address
4359 @var{addr} is the address where you want @value{GDBN} to begin displaying
4360 memory. The expression need not have a pointer value (though it may);
4361 it is always interpreted as an integer address of a byte of memory.
4362 @xref{Expressions, ,Expressions}, for more information on expressions. The default for
4363 @var{addr} is usually just after the last address examined---but several
4364 other commands also set the default address: @code{info breakpoints} (to
4365 the address of the last breakpoint listed), @code{info line} (to the
4366 starting address of a line), and @code{print} (if you use it to display
4367 a value from memory).
4370 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
4371 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
4372 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
4373 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
4374 @pxref{Registers}) in hexadecimal (@samp{x}).
4376 Since the letters indicating unit sizes are all distinct from the
4377 letters specifying output formats, you do not have to remember whether
4378 unit size or format comes first; either order works. The output
4379 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
4380 (However, the count @var{n} must come first; @samp{wx4} does not work.)
4382 Even though the unit size @var{u} is ignored for the formats @samp{s}
4383 and @samp{i}, you might still want to use a count @var{n}; for example,
4384 @samp{3i} specifies that you want to see three machine instructions,
4385 including any operands. The command @code{disassemble} gives an
4386 alternative way of inspecting machine instructions; @pxref{Machine
4387 Code,,Source and machine code}.
4389 All the defaults for the arguments to @code{x} are designed to make it
4390 easy to continue scanning memory with minimal specifications each time
4391 you use @code{x}. For example, after you have inspected three machine
4392 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
4393 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
4394 the repeat count @var{n} is used again; the other arguments default as
4395 for successive uses of @code{x}.
4397 @cindex @code{$_}, @code{$__}, and value history
4398 The addresses and contents printed by the @code{x} command are not saved
4399 in the value history because there is often too much of them and they
4400 would get in the way. Instead, @value{GDBN} makes these values available for
4401 subsequent use in expressions as values of the convenience variables
4402 @code{$_} and @code{$__}. After an @code{x} command, the last address
4403 examined is available for use in expressions in the convenience variable
4404 @code{$_}. The contents of that address, as examined, are available in
4405 the convenience variable @code{$__}.
4407 If the @code{x} command has a repeat count, the address and contents saved
4408 are from the last memory unit printed; this is not the same as the last
4409 address printed if several units were printed on the last line of output.
4412 @section Automatic display
4413 @cindex automatic display
4414 @cindex display of expressions
4416 If you find that you want to print the value of an expression frequently
4417 (to see how it changes), you might want to add it to the @dfn{automatic
4418 display list} so that @value{GDBN} prints its value each time your program stops.
4419 Each expression added to the list is given a number to identify it;
4420 to remove an expression from the list, you specify that number.
4421 The automatic display looks like this:
4425 3: bar[5] = (struct hack *) 0x3804
4429 This display shows item numbers, expressions and their current values. As with
4430 displays you request manually using @code{x} or @code{print}, you can
4431 specify the output format you prefer; in fact, @code{display} decides
4432 whether to use @code{print} or @code{x} depending on how elaborate your
4433 format specification is---it uses @code{x} if you specify a unit size,
4434 or one of the two formats (@samp{i} and @samp{s}) that are only
4435 supported by @code{x}; otherwise it uses @code{print}.
4439 @item display @var{exp}
4440 Add the expression @var{exp} to the list of expressions to display
4441 each time your program stops. @xref{Expressions, ,Expressions}.
4443 @code{display} does not repeat if you press @key{RET} again after using it.
4445 @item display/@var{fmt} @var{exp}
4446 For @var{fmt} specifying only a display format and not a size or
4447 count, add the expression @var{exp} to the auto-display list but
4448 arrange to display it each time in the specified format @var{fmt}.
4449 @xref{Output Formats,,Output formats}.
4451 @item display/@var{fmt} @var{addr}
4452 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
4453 number of units, add the expression @var{addr} as a memory address to
4454 be examined each time your program stops. Examining means in effect
4455 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory, ,Examining memory}.
4458 For example, @samp{display/i $pc} can be helpful, to see the machine
4459 instruction about to be executed each time execution stops (@samp{$pc}
4460 is a common name for the program counter; @pxref{Registers}).
4463 @kindex delete display
4465 @item undisplay @var{dnums}@dots{}
4466 @itemx delete display @var{dnums}@dots{}
4467 Remove item numbers @var{dnums} from the list of expressions to display.
4469 @code{undisplay} does not repeat if you press @key{RET} after using it.
4470 (Otherwise you would just get the error @samp{No display number @dots{}}.)
4472 @kindex disable display
4473 @item disable display @var{dnums}@dots{}
4474 Disable the display of item numbers @var{dnums}. A disabled display
4475 item is not printed automatically, but is not forgotten. It may be
4476 enabled again later.
4478 @kindex enable display
4479 @item enable display @var{dnums}@dots{}
4480 Enable display of item numbers @var{dnums}. It becomes effective once
4481 again in auto display of its expression, until you specify otherwise.
4484 Display the current values of the expressions on the list, just as is
4485 done when your program stops.
4487 @kindex info display
4489 Print the list of expressions previously set up to display
4490 automatically, each one with its item number, but without showing the
4491 values. This includes disabled expressions, which are marked as such.
4492 It also includes expressions which would not be displayed right now
4493 because they refer to automatic variables not currently available.
4496 If a display expression refers to local variables, then it does not make
4497 sense outside the lexical context for which it was set up. Such an
4498 expression is disabled when execution enters a context where one of its
4499 variables is not defined. For example, if you give the command
4500 @code{display last_char} while inside a function with an argument
4501 @code{last_char}, @value{GDBN} displays this argument while your program
4502 continues to stop inside that function. When it stops elsewhere---where
4503 there is no variable @code{last_char}---the display is disabled
4504 automatically. The next time your program stops where @code{last_char}
4505 is meaningful, you can enable the display expression once again.
4507 @node Print Settings
4508 @section Print settings
4510 @cindex format options
4511 @cindex print settings
4512 @value{GDBN} provides the following ways to control how arrays, structures,
4513 and symbols are printed.
4516 These settings are useful for debugging programs in any language:
4519 @kindex set print address
4520 @item set print address
4521 @itemx set print address on
4522 @value{GDBN} prints memory addresses showing the location of stack
4523 traces, structure values, pointer values, breakpoints, and so forth,
4524 even when it also displays the contents of those addresses. The default
4525 is @code{on}. For example, this is what a stack frame display looks like with
4526 @code{set print address on}:
4531 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
4533 530 if (lquote != def_lquote)
4537 @item set print address off
4538 Do not print addresses when displaying their contents. For example,
4539 this is the same stack frame displayed with @code{set print address off}:
4543 (@value{GDBP}) set print addr off
4545 #0 set_quotes (lq="<<", rq=">>") at input.c:530
4546 530 if (lquote != def_lquote)
4550 You can use @samp{set print address off} to eliminate all machine
4551 dependent displays from the @value{GDBN} interface. For example, with
4552 @code{print address off}, you should get the same text for backtraces on
4553 all machines---whether or not they involve pointer arguments.
4555 @kindex show print address
4556 @item show print address
4557 Show whether or not addresses are to be printed.
4560 When @value{GDBN} prints a symbolic address, it normally prints the
4561 closest earlier symbol plus an offset. If that symbol does not uniquely
4562 identify the address (for example, it is a name whose scope is a single
4563 source file), you may need to clarify. One way to do this is with
4564 @code{info line}, for example @samp{info line *0x4537}. Alternately,
4565 you can set @value{GDBN} to print the source file and line number when
4566 it prints a symbolic address:
4569 @kindex set print symbol-filename
4570 @item set print symbol-filename on
4571 Tell @value{GDBN} to print the source file name and line number of a
4572 symbol in the symbolic form of an address.
4574 @item set print symbol-filename off
4575 Do not print source file name and line number of a symbol. This is the
4578 @kindex show print symbol-filename
4579 @item show print symbol-filename
4580 Show whether or not @value{GDBN} will print the source file name and
4581 line number of a symbol in the symbolic form of an address.
4584 Another situation where it is helpful to show symbol filenames and line
4585 numbers is when disassembling code; @value{GDBN} shows you the line
4586 number and source file that corresponds to each instruction.
4588 Also, you may wish to see the symbolic form only if the address being
4589 printed is reasonably close to the closest earlier symbol:
4592 @kindex set print max-symbolic-offset
4593 @item set print max-symbolic-offset @var{max-offset}
4594 Tell @value{GDBN} to only display the symbolic form of an address if the
4595 offset between the closest earlier symbol and the address is less than
4596 @var{max-offset}. The default is 0, which tells @value{GDBN}
4597 to always print the symbolic form of an address if any symbol precedes it.
4599 @kindex show print max-symbolic-offset
4600 @item show print max-symbolic-offset
4601 Ask how large the maximum offset is that @value{GDBN} prints in a
4605 @cindex wild pointer, interpreting
4606 @cindex pointer, finding referent
4607 If you have a pointer and you are not sure where it points, try
4608 @samp{set print symbol-filename on}. Then you can determine the name
4609 and source file location of the variable where it points, using
4610 @samp{p/a @var{pointer}}. This interprets the address in symbolic form.
4611 For example, here @value{GDBN} shows that a variable @code{ptt} points
4612 at another variable @code{t}, defined in @file{hi2.c}:
4615 (@value{GDBP}) set print symbol-filename on
4616 (@value{GDBP}) p/a ptt
4617 $4 = 0xe008 <t in hi2.c>
4621 @emph{Warning:} For pointers that point to a local variable, @samp{p/a}
4622 does not show the symbol name and filename of the referent, even with
4623 the appropriate @code{set print} options turned on.
4626 Other settings control how different kinds of objects are printed:
4629 @kindex set print array
4630 @item set print array
4631 @itemx set print array on
4632 Pretty print arrays. This format is more convenient to read,
4633 but uses more space. The default is off.
4635 @item set print array off
4636 Return to compressed format for arrays.
4638 @kindex show print array
4639 @item show print array
4640 Show whether compressed or pretty format is selected for displaying
4643 @kindex set print elements
4644 @item set print elements @var{number-of-elements}
4645 Set a limit on how many elements of an array @value{GDBN} will print.
4646 If @value{GDBN} is printing a large array, it stops printing after it has
4647 printed the number of elements set by the @code{set print elements} command.
4648 This limit also applies to the display of strings.
4649 Setting @var{number-of-elements} to zero means that the printing is unlimited.
4651 @kindex show print elements
4652 @item show print elements
4653 Display the number of elements of a large array that @value{GDBN} will print.
4654 If the number is 0, then the printing is unlimited.
4656 @kindex set print null-stop
4657 @item set print null-stop
4658 Cause @value{GDBN} to stop printing the characters of an array when the first
4659 @sc{NULL} is encountered. This is useful when large arrays actually
4660 contain only short strings.
4662 @kindex set print pretty
4663 @item set print pretty on
4664 Cause @value{GDBN} to print structures in an indented format with one member
4665 per line, like this:
4680 @item set print pretty off
4681 Cause @value{GDBN} to print structures in a compact format, like this:
4685 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, \
4686 meat = 0x54 "Pork"@}
4691 This is the default format.
4693 @kindex show print pretty
4694 @item show print pretty
4695 Show which format @value{GDBN} is using to print structures.
4697 @kindex set print sevenbit-strings
4698 @item set print sevenbit-strings on
4699 Print using only seven-bit characters; if this option is set,
4700 @value{GDBN} displays any eight-bit characters (in strings or
4701 character values) using the notation @code{\}@var{nnn}. This setting is
4702 best if you are working in English (@sc{ascii}) and you use the
4703 high-order bit of characters as a marker or ``meta'' bit.
4705 @item set print sevenbit-strings off
4706 Print full eight-bit characters. This allows the use of more
4707 international character sets, and is the default.
4709 @kindex show print sevenbit-strings
4710 @item show print sevenbit-strings
4711 Show whether or not @value{GDBN} is printing only seven-bit characters.
4713 @kindex set print union
4714 @item set print union on
4715 Tell @value{GDBN} to print unions which are contained in structures. This
4716 is the default setting.
4718 @item set print union off
4719 Tell @value{GDBN} not to print unions which are contained in structures.
4721 @kindex show print union
4722 @item show print union
4723 Ask @value{GDBN} whether or not it will print unions which are contained in
4726 For example, given the declarations
4729 typedef enum @{Tree, Bug@} Species;
4730 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
4731 typedef enum @{Caterpillar, Cocoon, Butterfly@}
4742 struct thing foo = @{Tree, @{Acorn@}@};
4746 with @code{set print union on} in effect @samp{p foo} would print
4749 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
4753 and with @code{set print union off} in effect it would print
4756 $1 = @{it = Tree, form = @{...@}@}
4763 These settings are of interest when debugging C++ programs:
4767 @kindex set print demangle
4768 @item set print demangle
4769 @itemx set print demangle on
4770 Print C++ names in their source form rather than in the encoded
4771 (``mangled'') form passed to the assembler and linker for type-safe
4772 linkage. The default is @samp{on}.
4774 @kindex show print demangle
4775 @item show print demangle
4776 Show whether C++ names are printed in mangled or demangled form.
4778 @kindex set print asm-demangle
4779 @item set print asm-demangle
4780 @itemx set print asm-demangle on
4781 Print C++ names in their source form rather than their mangled form, even
4782 in assembler code printouts such as instruction disassemblies.
4785 @kindex show print asm-demangle
4786 @item show print asm-demangle
4787 Show whether C++ names in assembly listings are printed in mangled
4790 @kindex set demangle-style
4791 @cindex C++ symbol decoding style
4792 @cindex symbol decoding style, C++
4793 @item set demangle-style @var{style}
4794 Choose among several encoding schemes used by different compilers to
4795 represent C++ names. The choices for @var{style} are currently:
4799 Allow @value{GDBN} to choose a decoding style by inspecting your program.
4802 Decode based on the @sc{gnu} C++ compiler (@code{g++}) encoding algorithm.
4803 This is the default.
4806 Decode based on the Lucid C++ compiler (@code{lcc}) encoding algorithm.
4809 Decode using the algorithm in the @cite{C++ Annotated Reference Manual}.
4810 @strong{Warning:} this setting alone is not sufficient to allow
4811 debugging @code{cfront}-generated executables. @value{GDBN} would
4812 require further enhancement to permit that.
4815 Show the list of formats.
4818 @kindex show demangle-style
4819 @item show demangle-style
4820 Display the encoding style currently in use for decoding C++ symbols.
4822 @kindex set print object
4823 @item set print object
4824 @itemx set print object on
4825 When displaying a pointer to an object, identify the @emph{actual}
4826 (derived) type of the object rather than the @emph{declared} type, using
4827 the virtual function table.
4829 @item set print object off
4830 Display only the declared type of objects, without reference to the
4831 virtual function table. This is the default setting.
4833 @kindex show print object
4834 @item show print object
4835 Show whether actual, or declared, object types are displayed.
4837 @kindex set print static-members
4838 @item set print static-members
4839 @itemx set print static-members on
4840 Print static members when displaying a C++ object. The default is on.
4842 @item set print static-members off
4843 Do not print static members when displaying a C++ object.
4845 @kindex show print static-members
4846 @item show print static-members
4847 Show whether C++ static members are printed, or not.
4849 @kindex set print vtbl
4850 @item set print vtbl
4851 @itemx set print vtbl on
4852 Pretty print C++ virtual function tables. The default is off.
4854 @item set print vtbl off
4855 Do not pretty print C++ virtual function tables.
4857 @kindex show print vtbl
4858 @item show print vtbl
4859 Show whether C++ virtual function tables are pretty printed, or not.
4864 @section Value history
4866 @cindex value history
4867 Values printed by the @code{print} command are saved in the @value{GDBN}
4868 @dfn{value history}. This allows you to refer to them in other expressions.
4869 Values are kept until the symbol table is re-read or discarded
4870 (for example with the @code{file} or @code{symbol-file} commands).
4871 When the symbol table changes, the value history is discarded,
4872 since the values may contain pointers back to the types defined in the
4877 @cindex history number
4878 The values printed are given @dfn{history numbers} by which you can
4879 refer to them. These are successive integers starting with one.
4880 @code{print} shows you the history number assigned to a value by
4881 printing @samp{$@var{num} = } before the value; here @var{num} is the
4884 To refer to any previous value, use @samp{$} followed by the value's
4885 history number. The way @code{print} labels its output is designed to
4886 remind you of this. Just @code{$} refers to the most recent value in
4887 the history, and @code{$$} refers to the value before that.
4888 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
4889 is the value just prior to @code{$$}, @code{$$1} is equivalent to
4890 @code{$$}, and @code{$$0} is equivalent to @code{$}.
4892 For example, suppose you have just printed a pointer to a structure and
4893 want to see the contents of the structure. It suffices to type
4899 If you have a chain of structures where the component @code{next} points
4900 to the next one, you can print the contents of the next one with this:
4907 You can print successive links in the chain by repeating this
4908 command---which you can do by just typing @key{RET}.
4910 Note that the history records values, not expressions. If the value of
4911 @code{x} is 4 and you type these commands:
4919 then the value recorded in the value history by the @code{print} command
4920 remains 4 even though the value of @code{x} has changed.
4925 Print the last ten values in the value history, with their item numbers.
4926 This is like @samp{p@ $$9} repeated ten times, except that @code{show
4927 values} does not change the history.
4929 @item show values @var{n}
4930 Print ten history values centered on history item number @var{n}.
4933 Print ten history values just after the values last printed. If no more
4934 values are available, @code{show values +} produces no display.
4937 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
4938 same effect as @samp{show values +}.
4940 @node Convenience Vars
4941 @section Convenience variables
4943 @cindex convenience variables
4944 @value{GDBN} provides @dfn{convenience variables} that you can use within
4945 @value{GDBN} to hold on to a value and refer to it later. These variables
4946 exist entirely within @value{GDBN}; they are not part of your program, and
4947 setting a convenience variable has no direct effect on further execution
4948 of your program. That is why you can use them freely.
4950 Convenience variables are prefixed with @samp{$}. Any name preceded by
4951 @samp{$} can be used for a convenience variable, unless it is one of
4952 the predefined machine-specific register names (@pxref{Registers}).
4953 (Value history references, in contrast, are @emph{numbers} preceded
4954 by @samp{$}. @xref{Value History, ,Value history}.)
4956 You can save a value in a convenience variable with an assignment
4957 expression, just as you would set a variable in your program.
4961 set $foo = *object_ptr
4965 would save in @code{$foo} the value contained in the object pointed to by
4968 Using a convenience variable for the first time creates it, but its
4969 value is @code{void} until you assign a new value. You can alter the
4970 value with another assignment at any time.
4972 Convenience variables have no fixed types. You can assign a convenience
4973 variable any type of value, including structures and arrays, even if
4974 that variable already has a value of a different type. The convenience
4975 variable, when used as an expression, has the type of its current value.
4978 @kindex show convenience
4979 @item show convenience
4980 Print a list of convenience variables used so far, and their values.
4981 Abbreviated @code{show con}.
4984 One of the ways to use a convenience variable is as a counter to be
4985 incremented or a pointer to be advanced. For example, to print
4986 a field from successive elements of an array of structures:
4990 print bar[$i++]->contents
4993 @noindent Repeat that command by typing @key{RET}.
4995 Some convenience variables are created automatically by @value{GDBN} and given
4996 values likely to be useful.
5001 The variable @code{$_} is automatically set by the @code{x} command to
5002 the last address examined (@pxref{Memory, ,Examining memory}). Other
5003 commands which provide a default address for @code{x} to examine also
5004 set @code{$_} to that address; these commands include @code{info line}
5005 and @code{info breakpoint}. The type of @code{$_} is @code{void *}
5006 except when set by the @code{x} command, in which case it is a pointer
5007 to the type of @code{$__}.
5011 The variable @code{$__} is automatically set by the @code{x} command
5012 to the value found in the last address examined. Its type is chosen
5013 to match the format in which the data was printed.
5017 The variable @code{$_exitcode} is automatically set to the exit code when
5018 the program being debugged terminates.
5025 You can refer to machine register contents, in expressions, as variables
5026 with names starting with @samp{$}. The names of registers are different
5027 for each machine; use @code{info registers} to see the names used on
5031 @kindex info registers
5032 @item info registers
5033 Print the names and values of all registers except floating-point
5034 registers (in the selected stack frame).
5036 @kindex info all-registers
5037 @cindex floating point registers
5038 @item info all-registers
5039 Print the names and values of all registers, including floating-point
5042 @item info registers @var{regname} @dots{}
5043 Print the @dfn{relativized} value of each specified register @var{regname}.
5044 As discussed in detail below, register values are normally relative to
5045 the selected stack frame. @var{regname} may be any register name valid on
5046 the machine you are using, with or without the initial @samp{$}.
5049 @value{GDBN} has four ``standard'' register names that are available (in
5050 expressions) on most machines---whenever they do not conflict with an
5051 architecture's canonical mnemonics for registers. The register names
5052 @code{$pc} and @code{$sp} are used for the program counter register and
5053 the stack pointer. @code{$fp} is used for a register that contains a
5054 pointer to the current stack frame, and @code{$ps} is used for a
5055 register that contains the processor status. For example,
5056 you could print the program counter in hex with
5063 or print the instruction to be executed next with
5070 or add four to the stack pointer@footnote{This is a way of removing
5071 one word from the stack, on machines where stacks grow downward in
5072 memory (most machines, nowadays). This assumes that the innermost
5073 stack frame is selected; setting @code{$sp} is not allowed when other
5074 stack frames are selected. To pop entire frames off the stack,
5075 regardless of machine architecture, use @code{return};
5076 @pxref{Returning, ,Returning from a function}.} with
5082 Whenever possible, these four standard register names are available on
5083 your machine even though the machine has different canonical mnemonics,
5084 so long as there is no conflict. The @code{info registers} command
5085 shows the canonical names. For example, on the SPARC, @code{info
5086 registers} displays the processor status register as @code{$psr} but you
5087 can also refer to it as @code{$ps}.
5089 @value{GDBN} always considers the contents of an ordinary register as an
5090 integer when the register is examined in this way. Some machines have
5091 special registers which can hold nothing but floating point; these
5092 registers are considered to have floating point values. There is no way
5093 to refer to the contents of an ordinary register as floating point value
5094 (although you can @emph{print} it as a floating point value with
5095 @samp{print/f $@var{regname}}).
5097 Some registers have distinct ``raw'' and ``virtual'' data formats. This
5098 means that the data format in which the register contents are saved by
5099 the operating system is not the same one that your program normally
5100 sees. For example, the registers of the 68881 floating point
5101 coprocessor are always saved in ``extended'' (raw) format, but all C
5102 programs expect to work with ``double'' (virtual) format. In such
5103 cases, @value{GDBN} normally works with the virtual format only (the format
5104 that makes sense for your program), but the @code{info registers} command
5105 prints the data in both formats.
5107 Normally, register values are relative to the selected stack frame
5108 (@pxref{Selection, ,Selecting a frame}). This means that you get the
5109 value that the register would contain if all stack frames farther in
5110 were exited and their saved registers restored. In order to see the
5111 true contents of hardware registers, you must select the innermost
5112 frame (with @samp{frame 0}).
5114 However, @value{GDBN} must deduce where registers are saved, from the machine
5115 code generated by your compiler. If some registers are not saved, or if
5116 @value{GDBN} is unable to locate the saved registers, the selected stack
5117 frame makes no difference.
5121 @kindex set rstack_high_address
5122 @cindex AMD 29K register stack
5123 @cindex register stack, AMD29K
5124 @item set rstack_high_address @var{address}
5125 On AMD 29000 family processors, registers are saved in a separate
5126 ``register stack''. There is no way for @value{GDBN} to determine the extent
5127 of this stack. Normally, @value{GDBN} just assumes that the stack is ``large
5128 enough''. This may result in @value{GDBN} referencing memory locations that
5129 do not exist. If necessary, you can get around this problem by
5130 specifying the ending address of the register stack with the @code{set
5131 rstack_high_address} command. The argument should be an address, which
5132 you probably want to precede with @samp{0x} to specify in
5135 @kindex show rstack_high_address
5136 @item show rstack_high_address
5137 Display the current limit of the register stack, on AMD 29000 family
5143 @node Floating Point Hardware
5144 @section Floating point hardware
5145 @cindex floating point
5147 Depending on the configuration, @value{GDBN} may be able to give
5148 you more information about the status of the floating point hardware.
5153 Display hardware-dependent information about the floating
5154 point unit. The exact contents and layout vary depending on the
5155 floating point chip. Currently, @samp{info float} is supported on
5156 the ARM and x86 machines.
5162 @chapter Using @value{GDBN} with Different Languages
5166 Although programming languages generally have common aspects, they are
5167 rarely expressed in the same manner. For instance, in ANSI C,
5168 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
5169 Modula-2, it is accomplished by @code{p^}. Values can also be
5170 represented (and displayed) differently. Hex numbers in C appear as
5171 @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
5174 @cindex working language
5175 Language-specific information is built into @value{GDBN} for some languages,
5176 allowing you to express operations like the above in your program's
5177 native language, and allowing @value{GDBN} to output values in a manner
5178 consistent with the syntax of your program's native language. The
5179 language you use to build expressions is called the @dfn{working
5183 * Setting:: Switching between source languages
5184 * Show:: Displaying the language
5186 * Checks:: Type and range checks
5189 * Support:: Supported languages
5193 @section Switching between source languages
5195 There are two ways to control the working language---either have @value{GDBN}
5196 set it automatically, or select it manually yourself. You can use the
5197 @code{set language} command for either purpose. On startup, @value{GDBN}
5198 defaults to setting the language automatically. The working language is
5199 used to determine how expressions you type are interpreted, how values
5202 In addition to the working language, every source file that
5203 @value{GDBN} knows about has its own working language. For some object
5204 file formats, the compiler might indicate which language a particular
5205 source file is in. However, most of the time @value{GDBN} infers the
5206 language from the name of the file. The language of a source file
5207 controls whether C++ names are demangled---this way @code{backtrace} can
5208 show each frame appropriately for its own language. There is no way to
5209 set the language of a source file from within @value{GDBN}.
5211 This is most commonly a problem when you use a program, such
5212 as @code{cfront} or @code{f2c}, that generates C but is written in
5213 another language. In that case, make the
5214 program use @code{#line} directives in its C output; that way
5215 @value{GDBN} will know the correct language of the source code of the original
5216 program, and will display that source code, not the generated C code.
5219 * Filenames:: Filename extensions and languages.
5220 * Manually:: Setting the working language manually
5221 * Automatically:: Having @value{GDBN} infer the source language
5225 @subsection List of filename extensions and languages
5227 If a source file name ends in one of the following extensions, then
5228 @value{GDBN} infers that its language is the one indicated.
5233 Modula-2 source file
5254 Assembler source file. This actually behaves almost like C, but
5255 @value{GDBN} does not skip over function prologues when stepping.
5259 @subsection Setting the working language
5261 If you allow @value{GDBN} to set the language automatically,
5262 expressions are interpreted the same way in your debugging session and
5265 @kindex set language
5266 If you wish, you may set the language manually. To do this, issue the
5267 command @samp{set language @var{lang}}, where @var{lang} is the name of
5273 @code{c} or @code{modula-2}.
5275 For a list of the supported languages, type @samp{set language}.
5278 Setting the language manually prevents @value{GDBN} from updating the working
5279 language automatically. This can lead to confusion if you try
5280 to debug a program when the working language is not the same as the
5281 source language, when an expression is acceptable to both
5282 languages---but means different things. For instance, if the current
5283 source file were written in C, and @value{GDBN} was parsing Modula-2, a
5291 might not have the effect you intended. In C, this means to add
5292 @code{b} and @code{c} and place the result in @code{a}. The result
5293 printed would be the value of @code{a}. In Modula-2, this means to compare
5294 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
5298 @subsection Having @value{GDBN} infer the source language
5300 To have @value{GDBN} set the working language automatically, use
5301 @samp{set language local} or @samp{set language auto}. @value{GDBN}
5302 then infers the working language. That is, when your program stops in a
5303 frame (usually by encountering a breakpoint), @value{GDBN} sets the
5304 working language to the language recorded for the function in that
5305 frame. If the language for a frame is unknown (that is, if the function
5306 or block corresponding to the frame was defined in a source file that
5307 does not have a recognized extension), the current working language is
5308 not changed, and @value{GDBN} issues a warning.
5310 This may not seem necessary for most programs, which are written
5311 entirely in one source language. However, program modules and libraries
5312 written in one source language can be used by a main program written in
5313 a different source language. Using @samp{set language auto} in this
5314 case frees you from having to set the working language manually.
5317 @section Displaying the language
5319 The following commands help you find out which language is the
5320 working language, and also what language source files were written in.
5322 @kindex show language
5327 Display the current working language. This is the
5328 language you can use with commands such as @code{print} to
5329 build and compute expressions that may involve variables in your program.
5332 Display the source language for this frame. This language becomes the
5333 working language if you use an identifier from this frame.
5334 @xref{Frame Info, ,Information about a frame}, to identify the other
5335 information listed here.
5338 Display the source language of this source file.
5339 @xref{Symbols, ,Examining the Symbol Table}, to identify the other
5340 information listed here.
5345 @section Type and range checking
5348 @emph{Warning:} In this release, the @value{GDBN} commands for type and range
5349 checking are included, but they do not yet have any effect. This
5350 section documents the intended facilities.
5352 @c FIXME remove warning when type/range code added
5354 Some languages are designed to guard you against making seemingly common
5355 errors through a series of compile- and run-time checks. These include
5356 checking the type of arguments to functions and operators, and making
5357 sure mathematical overflows are caught at run time. Checks such as
5358 these help to ensure a program's correctness once it has been compiled
5359 by eliminating type mismatches, and providing active checks for range
5360 errors when your program is running.
5362 @value{GDBN} can check for conditions like the above if you wish.
5363 Although @value{GDBN} does not check the statements in your program, it
5364 can check expressions entered directly into @value{GDBN} for evaluation via
5365 the @code{print} command, for example. As with the working language,
5366 @value{GDBN} can also decide whether or not to check automatically based on
5367 your program's source language. @xref{Support, ,Supported languages},
5368 for the default settings of supported languages.
5371 * Type Checking:: An overview of type checking
5372 * Range Checking:: An overview of range checking
5375 @cindex type checking
5376 @cindex checks, type
5378 @subsection An overview of type checking
5380 Some languages, such as Modula-2, are strongly typed, meaning that the
5381 arguments to operators and functions have to be of the correct type,
5382 otherwise an error occurs. These checks prevent type mismatch
5383 errors from ever causing any run-time problems. For example,
5391 The second example fails because the @code{CARDINAL} 1 is not
5392 type-compatible with the @code{REAL} 2.3.
5394 For the expressions you use in @value{GDBN} commands, you can tell the
5395 @value{GDBN} type checker to skip checking;
5396 to treat any mismatches as errors and abandon the expression;
5397 or to only issue warnings when type mismatches occur,
5398 but evaluate the expression anyway. When you choose the last of
5399 these, @value{GDBN} evaluates expressions like the second example above, but
5400 also issues a warning.
5402 Even if you turn type checking off, there may be other reasons
5403 related to type that prevent @value{GDBN} from evaluating an expression.
5404 For instance, @value{GDBN} does not know how to add an @code{int} and
5405 a @code{struct foo}. These particular type errors have nothing to do
5406 with the language in use, and usually arise from expressions, such as
5407 the one described above, which make little sense to evaluate anyway.
5409 Each language defines to what degree it is strict about type. For
5410 instance, both Modula-2 and C require the arguments to arithmetical
5411 operators to be numbers. In C, enumerated types and pointers can be
5412 represented as numbers, so that they are valid arguments to mathematical
5413 operators. @xref{Support, ,Supported languages}, for further
5414 details on specific languages.
5416 @value{GDBN} provides some additional commands for controlling the type checker:
5419 @kindex set check type
5420 @kindex show check type
5422 @item set check type auto
5423 Set type checking on or off based on the current working language.
5424 @xref{Support, ,Supported languages}, for the default settings for
5427 @item set check type on
5428 @itemx set check type off
5429 Set type checking on or off, overriding the default setting for the
5430 current working language. Issue a warning if the setting does not
5431 match the language default. If any type mismatches occur in
5432 evaluating an expression while typechecking is on, @value{GDBN} prints a
5433 message and aborts evaluation of the expression.
5435 @item set check type warn
5436 Cause the type checker to issue warnings, but to always attempt to
5437 evaluate the expression. Evaluating the expression may still
5438 be impossible for other reasons. For example, @value{GDBN} cannot add
5439 numbers and structures.
5442 Show the current setting of the type checker, and whether or not @value{GDBN}
5443 is setting it automatically.
5446 @cindex range checking
5447 @cindex checks, range
5448 @node Range Checking
5449 @subsection An overview of range checking
5451 In some languages (such as Modula-2), it is an error to exceed the
5452 bounds of a type; this is enforced with run-time checks. Such range
5453 checking is meant to ensure program correctness by making sure
5454 computations do not overflow, or indices on an array element access do
5455 not exceed the bounds of the array.
5457 For expressions you use in @value{GDBN} commands, you can tell
5458 @value{GDBN} to treat range errors in one of three ways: ignore them,
5459 always treat them as errors and abandon the expression, or issue
5460 warnings but evaluate the expression anyway.
5462 A range error can result from numerical overflow, from exceeding an
5463 array index bound, or when you type a constant that is not a member
5464 of any type. Some languages, however, do not treat overflows as an
5465 error. In many implementations of C, mathematical overflow causes the
5466 result to ``wrap around'' to lower values---for example, if @var{m} is
5467 the largest integer value, and @var{s} is the smallest, then
5470 @var{m} + 1 @result{} @var{s}
5473 This, too, is specific to individual languages, and in some cases
5474 specific to individual compilers or machines. @xref{Support, ,
5475 Supported languages}, for further details on specific languages.
5477 @value{GDBN} provides some additional commands for controlling the range checker:
5480 @kindex set check range
5481 @kindex show check range
5483 @item set check range auto
5484 Set range checking on or off based on the current working language.
5485 @xref{Support, ,Supported languages}, for the default settings for
5488 @item set check range on
5489 @itemx set check range off
5490 Set range checking on or off, overriding the default setting for the
5491 current working language. A warning is issued if the setting does not
5492 match the language default. If a range error occurs, then a message
5493 is printed and evaluation of the expression is aborted.
5495 @item set check range warn
5496 Output messages when the @value{GDBN} range checker detects a range error,
5497 but attempt to evaluate the expression anyway. Evaluating the
5498 expression may still be impossible for other reasons, such as accessing
5499 memory that the process does not own (a typical example from many Unix
5503 Show the current setting of the range checker, and whether or not it is
5504 being set automatically by @value{GDBN}.
5509 @section Supported languages
5512 @value{GDBN} 4 supports C, C++, and Modula-2.
5515 @value{GDBN} 4 supports C, and C++.
5517 Some @value{GDBN} features may be used in expressions regardless of the
5518 language you use: the @value{GDBN} @code{@@} and @code{::} operators,
5519 and the @samp{@{type@}addr} construct (@pxref{Expressions,
5520 ,Expressions}) can be used with the constructs of any supported
5523 The following sections detail to what degree each source language is
5524 supported by @value{GDBN}. These sections are not meant to be language
5525 tutorials or references, but serve only as a reference guide to what the
5526 @value{GDBN} expression parser accepts, and what input and output
5527 formats should look like for different languages. There are many good
5528 books written on each of these languages; please look to these for a
5529 language reference or tutorial.
5534 * Modula-2:: Modula-2
5538 @subsection C and C++
5540 @cindex expressions in C or C++
5542 Since C and C++ are so closely related, many features of @value{GDBN} apply
5543 to both languages. Whenever this is the case, we discuss those languages
5547 @c Cancel this below, under same condition, at end of this chapter!
5553 @cindex @sc{gnu} C++
5554 The C++ debugging facilities are jointly implemented by the @sc{gnu} C++
5555 compiler and @value{GDBN}. Therefore, to debug your C++ code
5556 effectively, you must compile your C++ programs with the @sc{gnu} C++
5557 compiler, @code{g++}.
5559 For best results when debugging C++ programs, use the stabs debugging
5560 format. You can select that format explicitly with the @code{g++}
5561 command-line options @samp{-gstabs} or @samp{-gstabs+}. See
5562 @ref{Debugging Options,,Options for Debugging Your Program or @sc{gnu} CC,
5563 gcc.info, Using @sc{gnu} CC}, for more information.
5567 @chapter C Language Support
5569 @cindex expressions in C
5571 Information specific to the C language is built into @value{GDBN} so that you
5572 can use C expressions while degugging. This also permits @value{GDBN} to
5573 output values in a manner consistent with C conventions.
5576 * C Operators:: C operators
5577 * C Constants:: C constants
5578 * Debugging C:: @value{GDBN} and C
5583 * C Operators:: C and C++ operators
5584 * C Constants:: C and C++ constants
5585 * Cplus expressions:: C++ expressions
5586 * C Defaults:: Default settings for C and C++
5588 * C Checks:: C and C++ type and range checks
5591 * Debugging C:: @value{GDBN} and C
5592 * Debugging C plus plus:: Special features for C++
5597 @cindex C and C++ operators
5599 @subsubsection C and C++ operators
5604 @section C operators
5607 Operators must be defined on values of specific types. For instance,
5608 @code{+} is defined on numbers, but not on structures. Operators are
5609 often defined on groups of types.
5612 For the purposes of C and C++, the following definitions hold:
5617 @emph{Integral types} include @code{int} with any of its storage-class
5618 specifiers; @code{char}; and @code{enum}.
5621 @emph{Floating-point types} include @code{float} and @code{double}.
5624 @emph{Pointer types} include all types defined as @code{(@var{type}
5628 @emph{Scalar types} include all of the above.
5632 The following operators are supported. They are listed here
5633 in order of increasing precedence:
5637 The comma or sequencing operator. Expressions in a comma-separated list
5638 are evaluated from left to right, with the result of the entire
5639 expression being the last expression evaluated.
5642 Assignment. The value of an assignment expression is the value
5643 assigned. Defined on scalar types.
5646 Used in an expression of the form @w{@code{@var{a} @var{op}= @var{b}}},
5647 and translated to @w{@code{@var{a} = @var{a op b}}}.
5648 @w{@code{@var{op}=}} and @code{=} have the same precendence.
5649 @var{op} is any one of the operators @code{|}, @code{^}, @code{&},
5650 @code{<<}, @code{>>}, @code{+}, @code{-}, @code{*}, @code{/}, @code{%}.
5653 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
5654 of as: if @var{a} then @var{b} else @var{c}. @var{a} should be of an
5658 Logical @sc{or}. Defined on integral types.
5661 Logical @sc{and}. Defined on integral types.
5664 Bitwise @sc{or}. Defined on integral types.
5667 Bitwise exclusive-@sc{or}. Defined on integral types.
5670 Bitwise @sc{and}. Defined on integral types.
5673 Equality and inequality. Defined on scalar types. The value of these
5674 expressions is 0 for false and non-zero for true.
5676 @item <@r{, }>@r{, }<=@r{, }>=
5677 Less than, greater than, less than or equal, greater than or equal.
5678 Defined on scalar types. The value of these expressions is 0 for false
5679 and non-zero for true.
5682 left shift, and right shift. Defined on integral types.
5685 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
5688 Addition and subtraction. Defined on integral types, floating-point types and
5691 @item *@r{, }/@r{, }%
5692 Multiplication, division, and modulus. Multiplication and division are
5693 defined on integral and floating-point types. Modulus is defined on
5697 Increment and decrement. When appearing before a variable, the
5698 operation is performed before the variable is used in an expression;
5699 when appearing after it, the variable's value is used before the
5700 operation takes place.
5703 Pointer dereferencing. Defined on pointer types. Same precedence as
5707 Address operator. Defined on variables. Same precedence as @code{++}.
5710 For debugging C++, @value{GDBN} implements a use of @samp{&} beyond what is
5711 allowed in the C++ language itself: you can use @samp{&(&@var{ref})}
5712 (or, if you prefer, simply @samp{&&@var{ref}}) to examine the address
5713 where a C++ reference variable (declared with @samp{&@var{ref}}) is
5718 Negative. Defined on integral and floating-point types. Same
5719 precedence as @code{++}.
5722 Logical negation. Defined on integral types. Same precedence as
5726 Bitwise complement operator. Defined on integral types. Same precedence as
5731 Structure member, and pointer-to-structure member. For convenience,
5732 @value{GDBN} regards the two as equivalent, choosing whether to dereference a
5733 pointer based on the stored type information.
5734 Defined on @code{struct} and @code{union} data.
5737 Array indexing. @code{@var{a}[@var{i}]} is defined as
5738 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
5741 Function parameter list. Same precedence as @code{->}.
5745 C++ scope resolution operator. Defined on
5746 @code{struct}, @code{union}, and @code{class} types.
5754 represent the @value{GDBN} scope operator (@pxref{Expressions,
5757 Same precedence as @code{::}, above.
5762 @cindex C and C++ constants
5764 @subsubsection C and C++ constants
5766 @value{GDBN} allows you to express the constants of C and C++ in the
5772 @section C constants
5774 @value{GDBN} allows you to express the constants of C in the
5780 Integer constants are a sequence of digits. Octal constants are
5781 specified by a leading @samp{0} (i.e. zero), and hexadecimal constants by
5782 a leading @samp{0x} or @samp{0X}. Constants may also end with a letter
5783 @samp{l}, specifying that the constant should be treated as a
5787 Floating point constants are a sequence of digits, followed by a decimal
5788 point, followed by a sequence of digits, and optionally followed by an
5789 exponent. An exponent is of the form:
5790 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
5791 sequence of digits. The @samp{+} is optional for positive exponents.
5794 Enumerated constants consist of enumerated identifiers, or their
5795 integral equivalents.
5798 Character constants are a single character surrounded by single quotes
5799 (@code{'}), or a number---the ordinal value of the corresponding character
5800 (usually its @sc{ASCII} value). Within quotes, the single character may
5801 be represented by a letter or by @dfn{escape sequences}, which are of
5802 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
5803 of the character's ordinal value; or of the form @samp{\@var{x}}, where
5804 @samp{@var{x}} is a predefined special character---for example,
5805 @samp{\n} for newline.
5808 String constants are a sequence of character constants surrounded
5809 by double quotes (@code{"}).
5812 Pointer constants are an integral value. You can also write pointers
5813 to constants using the C operator @samp{&}.
5816 Array constants are comma-separated lists surrounded by braces @samp{@{}
5817 and @samp{@}}; for example, @samp{@{1,2,3@}} is a three-element array of
5818 integers, @samp{@{@{1,2@}, @{3,4@}, @{5,6@}@}} is a three-by-two array,
5819 and @samp{@{&"hi", &"there", &"fred"@}} is a three-element array of pointers.
5823 @node Cplus expressions
5824 @subsubsection C++ expressions
5826 @cindex expressions in C++
5827 @value{GDBN} expression handling has a number of extensions to
5828 interpret a significant subset of C++ expressions.
5830 @cindex C++ support, not in @sc{coff}
5831 @cindex @sc{coff} versus C++
5832 @cindex C++ and object formats
5833 @cindex object formats and C++
5834 @cindex a.out and C++
5835 @cindex @sc{ecoff} and C++
5836 @cindex @sc{xcoff} and C++
5837 @cindex @sc{elf}/stabs and C++
5838 @cindex @sc{elf}/@sc{dwarf} and C++
5839 @c FIXME!! GDB may eventually be able to debug C++ using DWARF; check
5840 @c periodically whether this has happened...
5842 @emph{Warning:} @value{GDBN} can only debug C++ code if you compile with
5843 the @sc{gnu} C++ compiler. Moreover, C++ debugging depends on the use of
5844 additional debugging information in the symbol table, and thus requires
5845 special support. @value{GDBN} has this support @emph{only} with the
5846 stabs debug format. In particular, if your compiler generates a.out,
5847 MIPS @sc{ecoff}, RS/6000 @sc{xcoff}, or @sc{elf} with stabs extensions
5848 to the symbol table, these facilities are all available. (With @sc{gnu} CC,
5849 you can use the @samp{-gstabs} option to request stabs debugging
5850 extensions explicitly.) Where the object code format is standard
5851 @sc{coff} or @sc{dwarf} in @sc{elf}, on the other hand, most of the C++
5852 support in @value{GDBN} does @emph{not} work.
5857 @cindex member functions
5859 Member function calls are allowed; you can use expressions like
5862 count = aml->GetOriginal(x, y)
5866 @cindex namespace in C++
5868 While a member function is active (in the selected stack frame), your
5869 expressions have the same namespace available as the member function;
5870 that is, @value{GDBN} allows implicit references to the class instance
5871 pointer @code{this} following the same rules as C++.
5873 @cindex call overloaded functions
5874 @cindex type conversions in C++
5876 You can call overloaded functions; @value{GDBN} resolves the function
5877 call to the right definition, with one restriction---you must use
5878 arguments of the type required by the function that you want to call.
5879 @value{GDBN} does not perform conversions requiring constructors or
5880 user-defined type operators.
5882 @cindex reference declarations
5884 @value{GDBN} understands variables declared as C++ references; you can use
5885 them in expressions just as you do in C++ source---they are automatically
5888 In the parameter list shown when @value{GDBN} displays a frame, the values of
5889 reference variables are not displayed (unlike other variables); this
5890 avoids clutter, since references are often used for large structures.
5891 The @emph{address} of a reference variable is always shown, unless
5892 you have specified @samp{set print address off}.
5895 @value{GDBN} supports the C++ name resolution operator @code{::}---your
5896 expressions can use it just as expressions in your program do. Since
5897 one scope may be defined in another, you can use @code{::} repeatedly if
5898 necessary, for example in an expression like
5899 @samp{@var{scope1}::@var{scope2}::@var{name}}. @value{GDBN} also allows
5900 resolving name scope by reference to source files, in both C and C++
5901 debugging (@pxref{Variables, ,Program variables}).
5905 @subsubsection C and C++ defaults
5906 @cindex C and C++ defaults
5908 If you allow @value{GDBN} to set type and range checking automatically, they
5909 both default to @code{off} whenever the working language changes to
5910 C or C++. This happens regardless of whether you or @value{GDBN}
5911 selects the working language.
5913 If you allow @value{GDBN} to set the language automatically, it recognizes
5914 source files whose names end with @file{.c}, @file{.C}, or @file{.cc}, and
5915 when @value{GDBN} enters code compiled from one of these files,
5916 it sets the working language to C or C++.
5917 @xref{Automatically, ,Having @value{GDBN} infer the source language}, for
5921 @c Type checking is (a) primarily motivated by Modula-2, and (b)
5922 @c unimplemented. If (b) changes, it might make sense to let this node
5923 @c appear even if Mod-2 does not, but meanwhile ignore it. roland 16jul93.
5925 @subsubsection C and C++ type and range checks
5926 @cindex C and C++ checks
5928 By default, when @value{GDBN} parses C or C++ expressions, type checking
5929 is not used. However, if you turn type checking on, @value{GDBN}
5930 considers two variables type equivalent if:
5934 The two variables are structured and have the same structure, union, or
5938 The two variables have the same type name, or types that have been
5939 declared equivalent through @code{typedef}.
5942 @c leaving this out because neither J Gilmore nor R Pesch understand it.
5945 The two @code{struct}, @code{union}, or @code{enum} variables are
5946 declared in the same declaration. (Note: this may not be true for all C
5951 Range checking, if turned on, is done on mathematical operations. Array
5952 indices are not checked, since they are often used to index a pointer
5953 that is not itself an array.
5959 @subsubsection @value{GDBN} and C
5963 @section @value{GDBN} and C
5966 The @code{set print union} and @code{show print union} commands apply to
5967 the @code{union} type. When set to @samp{on}, any @code{union} that is
5968 inside a @code{struct}
5973 Otherwise, it appears as @samp{@{...@}}.
5975 The @code{@@} operator aids in the debugging of dynamic arrays, formed
5976 with pointers and a memory allocation function. @xref{Expressions,
5980 @node Debugging C plus plus
5981 @subsubsection @value{GDBN} features for C++
5983 @cindex commands for C++
5984 Some @value{GDBN} commands are particularly useful with C++, and some are
5985 designed specifically for use with C++. Here is a summary:
5988 @cindex break in overloaded functions
5989 @item @r{breakpoint menus}
5990 When you want a breakpoint in a function whose name is overloaded,
5991 @value{GDBN} breakpoint menus help you specify which function definition
5992 you want. @xref{Breakpoint Menus,,Breakpoint menus}.
5994 @cindex overloading in C++
5995 @item rbreak @var{regex}
5996 Setting breakpoints using regular expressions is helpful for setting
5997 breakpoints on overloaded functions that are not members of any special
5999 @xref{Set Breaks, ,Setting breakpoints}.
6001 @cindex C++ exception handling
6002 @item catch @var{exceptions}
6004 Debug C++ exception handling using these commands. @xref{Exception
6005 Handling, ,Breakpoints and exceptions}.
6008 @item ptype @var{typename}
6009 Print inheritance relationships as well as other information for type
6011 @xref{Symbols, ,Examining the Symbol Table}.
6013 @cindex C++ symbol display
6014 @item set print demangle
6015 @itemx show print demangle
6016 @itemx set print asm-demangle
6017 @itemx show print asm-demangle
6018 Control whether C++ symbols display in their source form, both when
6019 displaying code as C++ source and when displaying disassemblies.
6020 @xref{Print Settings, ,Print settings}.
6022 @item set print object
6023 @itemx show print object
6024 Choose whether to print derived (actual) or declared types of objects.
6025 @xref{Print Settings, ,Print settings}.
6027 @item set print vtbl
6028 @itemx show print vtbl
6029 Control the format for printing virtual function tables.
6030 @xref{Print Settings, ,Print settings}.
6032 @item @r{Overloaded symbol names}
6033 You can specify a particular definition of an overloaded symbol, using
6034 the same notation that is used to declare such symbols in C++: type
6035 @code{@var{symbol}(@var{types})} rather than just @var{symbol}. You can
6036 also use the @value{GDBN} command-line word completion facilities to list the
6037 available choices, or to finish the type list for you.
6038 @xref{Completion,, Command completion}, for details on how to do this.
6041 @c cancels "raisesections" under same conditions near bgn of chapter
6047 @subsection Modula-2
6050 The extensions made to @value{GDBN} to support Modula-2 only support
6051 output from the @sc{gnu} Modula-2 compiler (which is currently being
6052 developed). Other Modula-2 compilers are not currently supported, and
6053 attempting to debug executables produced by them is most likely
6054 to give an error as @value{GDBN} reads in the executable's symbol
6057 @cindex expressions in Modula-2
6059 * M2 Operators:: Built-in operators
6060 * Built-In Func/Proc:: Built-in functions and procedures
6061 * M2 Constants:: Modula-2 constants
6062 * M2 Defaults:: Default settings for Modula-2
6063 * Deviations:: Deviations from standard Modula-2
6064 * M2 Checks:: Modula-2 type and range checks
6065 * M2 Scope:: The scope operators @code{::} and @code{.}
6066 * GDB/M2:: @value{GDBN} and Modula-2
6070 @subsubsection Operators
6071 @cindex Modula-2 operators
6073 Operators must be defined on values of specific types. For instance,
6074 @code{+} is defined on numbers, but not on structures. Operators are
6075 often defined on groups of types. For the purposes of Modula-2, the
6076 following definitions hold:
6081 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
6085 @emph{Character types} consist of @code{CHAR} and its subranges.
6088 @emph{Floating-point types} consist of @code{REAL}.
6091 @emph{Pointer types} consist of anything declared as @code{POINTER TO
6095 @emph{Scalar types} consist of all of the above.
6098 @emph{Set types} consist of @code{SET} and @code{BITSET} types.
6101 @emph{Boolean types} consist of @code{BOOLEAN}.
6105 The following operators are supported, and appear in order of
6106 increasing precedence:
6110 Function argument or array index separator.
6113 Assignment. The value of @var{var} @code{:=} @var{value} is
6117 Less than, greater than on integral, floating-point, or enumerated
6121 Less than, greater than, less than or equal to, greater than or equal to
6122 on integral, floating-point and enumerated types, or set inclusion on
6123 set types. Same precedence as @code{<}.
6125 @item =@r{, }<>@r{, }#
6126 Equality and two ways of expressing inequality, valid on scalar types.
6127 Same precedence as @code{<}. In @value{GDBN} scripts, only @code{<>} is
6128 available for inequality, since @code{#} conflicts with the script
6132 Set membership. Defined on set types and the types of their members.
6133 Same precedence as @code{<}.
6136 Boolean disjunction. Defined on boolean types.
6139 Boolean conjuction. Defined on boolean types.
6142 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
6145 Addition and subtraction on integral and floating-point types, or union
6146 and difference on set types.
6149 Multiplication on integral and floating-point types, or set intersection
6153 Division on floating-point types, or symmetric set difference on set
6154 types. Same precedence as @code{*}.
6157 Integer division and remainder. Defined on integral types. Same
6158 precedence as @code{*}.
6161 Negative. Defined on @code{INTEGER} and @code{REAL} data.
6164 Pointer dereferencing. Defined on pointer types.
6167 Boolean negation. Defined on boolean types. Same precedence as
6171 @code{RECORD} field selector. Defined on @code{RECORD} data. Same
6172 precedence as @code{^}.
6175 Array indexing. Defined on @code{ARRAY} data. Same precedence as @code{^}.
6178 Procedure argument list. Defined on @code{PROCEDURE} objects. Same precedence
6182 @value{GDBN} and Modula-2 scope operators.
6186 @emph{Warning:} Sets and their operations are not yet supported, so @value{GDBN}
6187 treats the use of the operator @code{IN}, or the use of operators
6188 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
6189 @code{<=}, and @code{>=} on sets as an error.
6192 @cindex Modula-2 built-ins
6193 @node Built-In Func/Proc
6194 @subsubsection Built-in functions and procedures
6196 Modula-2 also makes available several built-in procedures and functions.
6197 In describing these, the following metavariables are used:
6202 represents an @code{ARRAY} variable.
6205 represents a @code{CHAR} constant or variable.
6208 represents a variable or constant of integral type.
6211 represents an identifier that belongs to a set. Generally used in the
6212 same function with the metavariable @var{s}. The type of @var{s} should
6213 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}).
6216 represents a variable or constant of integral or floating-point type.
6219 represents a variable or constant of floating-point type.
6225 represents a variable.
6228 represents a variable or constant of one of many types. See the
6229 explanation of the function for details.
6232 All Modula-2 built-in procedures also return a result, described below.
6236 Returns the absolute value of @var{n}.
6239 If @var{c} is a lower case letter, it returns its upper case
6240 equivalent, otherwise it returns its argument
6243 Returns the character whose ordinal value is @var{i}.
6246 Decrements the value in the variable @var{v}. Returns the new value.
6248 @item DEC(@var{v},@var{i})
6249 Decrements the value in the variable @var{v} by @var{i}. Returns the
6252 @item EXCL(@var{m},@var{s})
6253 Removes the element @var{m} from the set @var{s}. Returns the new
6256 @item FLOAT(@var{i})
6257 Returns the floating point equivalent of the integer @var{i}.
6260 Returns the index of the last member of @var{a}.
6263 Increments the value in the variable @var{v}. Returns the new value.
6265 @item INC(@var{v},@var{i})
6266 Increments the value in the variable @var{v} by @var{i}. Returns the
6269 @item INCL(@var{m},@var{s})
6270 Adds the element @var{m} to the set @var{s} if it is not already
6271 there. Returns the new set.
6274 Returns the maximum value of the type @var{t}.
6277 Returns the minimum value of the type @var{t}.
6280 Returns boolean TRUE if @var{i} is an odd number.
6283 Returns the ordinal value of its argument. For example, the ordinal
6284 value of a character is its ASCII value (on machines supporting the
6285 ASCII character set). @var{x} must be of an ordered type, which include
6286 integral, character and enumerated types.
6289 Returns the size of its argument. @var{x} can be a variable or a type.
6291 @item TRUNC(@var{r})
6292 Returns the integral part of @var{r}.
6294 @item VAL(@var{t},@var{i})
6295 Returns the member of the type @var{t} whose ordinal value is @var{i}.
6299 @emph{Warning:} Sets and their operations are not yet supported, so
6300 @value{GDBN} treats the use of procedures @code{INCL} and @code{EXCL} as
6304 @cindex Modula-2 constants
6306 @subsubsection Constants
6308 @value{GDBN} allows you to express the constants of Modula-2 in the following
6314 Integer constants are simply a sequence of digits. When used in an
6315 expression, a constant is interpreted to be type-compatible with the
6316 rest of the expression. Hexadecimal integers are specified by a
6317 trailing @samp{H}, and octal integers by a trailing @samp{B}.
6320 Floating point constants appear as a sequence of digits, followed by a
6321 decimal point and another sequence of digits. An optional exponent can
6322 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
6323 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
6324 digits of the floating point constant must be valid decimal (base 10)
6328 Character constants consist of a single character enclosed by a pair of
6329 like quotes, either single (@code{'}) or double (@code{"}). They may
6330 also be expressed by their ordinal value (their ASCII value, usually)
6331 followed by a @samp{C}.
6334 String constants consist of a sequence of characters enclosed by a
6335 pair of like quotes, either single (@code{'}) or double (@code{"}).
6336 Escape sequences in the style of C are also allowed. @xref{C
6337 Constants, ,C and C++ constants}, for a brief explanation of escape
6341 Enumerated constants consist of an enumerated identifier.
6344 Boolean constants consist of the identifiers @code{TRUE} and
6348 Pointer constants consist of integral values only.
6351 Set constants are not yet supported.
6355 @subsubsection Modula-2 defaults
6356 @cindex Modula-2 defaults
6358 If type and range checking are set automatically by @value{GDBN}, they
6359 both default to @code{on} whenever the working language changes to
6360 Modula-2. This happens regardless of whether you, or @value{GDBN},
6361 selected the working language.
6363 If you allow @value{GDBN} to set the language automatically, then entering
6364 code compiled from a file whose name ends with @file{.mod} sets the
6365 working language to Modula-2. @xref{Automatically, ,Having @value{GDBN} set
6366 the language automatically}, for further details.
6369 @subsubsection Deviations from standard Modula-2
6370 @cindex Modula-2, deviations from
6372 A few changes have been made to make Modula-2 programs easier to debug.
6373 This is done primarily via loosening its type strictness:
6377 Unlike in standard Modula-2, pointer constants can be formed by
6378 integers. This allows you to modify pointer variables during
6379 debugging. (In standard Modula-2, the actual address contained in a
6380 pointer variable is hidden from you; it can only be modified
6381 through direct assignment to another pointer variable or expression that
6382 returned a pointer.)
6385 C escape sequences can be used in strings and characters to represent
6386 non-printable characters. @value{GDBN} prints out strings with these
6387 escape sequences embedded. Single non-printable characters are
6388 printed using the @samp{CHR(@var{nnn})} format.
6391 The assignment operator (@code{:=}) returns the value of its right-hand
6395 All built-in procedures both modify @emph{and} return their argument.
6399 @subsubsection Modula-2 type and range checks
6400 @cindex Modula-2 checks
6403 @emph{Warning:} in this release, @value{GDBN} does not yet perform type or
6406 @c FIXME remove warning when type/range checks added
6408 @value{GDBN} considers two Modula-2 variables type equivalent if:
6412 They are of types that have been declared equivalent via a @code{TYPE
6413 @var{t1} = @var{t2}} statement
6416 They have been declared on the same line. (Note: This is true of the
6417 @sc{gnu} Modula-2 compiler, but it may not be true of other compilers.)
6420 As long as type checking is enabled, any attempt to combine variables
6421 whose types are not equivalent is an error.
6423 Range checking is done on all mathematical operations, assignment, array
6424 index bounds, and all built-in functions and procedures.
6427 @subsubsection The scope operators @code{::} and @code{.}
6430 @cindex colon, doubled as scope operator
6433 @c Info cannot handle :: but TeX can.
6439 There are a few subtle differences between the Modula-2 scope operator
6440 (@code{.}) and the @value{GDBN} scope operator (@code{::}). The two have
6445 @var{module} . @var{id}
6446 @var{scope} :: @var{id}
6450 where @var{scope} is the name of a module or a procedure,
6451 @var{module} the name of a module, and @var{id} is any declared
6452 identifier within your program, except another module.
6454 Using the @code{::} operator makes @value{GDBN} search the scope
6455 specified by @var{scope} for the identifier @var{id}. If it is not
6456 found in the specified scope, then @value{GDBN} searches all scopes
6457 enclosing the one specified by @var{scope}.
6459 Using the @code{.} operator makes @value{GDBN} search the current scope for
6460 the identifier specified by @var{id} that was imported from the
6461 definition module specified by @var{module}. With this operator, it is
6462 an error if the identifier @var{id} was not imported from definition
6463 module @var{module}, or if @var{id} is not an identifier in
6467 @subsubsection @value{GDBN} and Modula-2
6469 Some @value{GDBN} commands have little use when debugging Modula-2 programs.
6470 Five subcommands of @code{set print} and @code{show print} apply
6471 specifically to C and C++: @samp{vtbl}, @samp{demangle},
6472 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
6473 apply to C++, and the last to the C @code{union} type, which has no direct
6474 analogue in Modula-2.
6476 The @code{@@} operator (@pxref{Expressions, ,Expressions}), while available
6477 while using any language, is not useful with Modula-2. Its
6478 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
6479 created in Modula-2 as they can in C or C++. However, because an
6480 address can be specified by an integral constant, the construct
6481 @samp{@{@var{type}@}@var{adrexp}} is still useful. (@pxref{Expressions, ,Expressions})
6483 @cindex @code{#} in Modula-2
6484 In @value{GDBN} scripts, the Modula-2 inequality operator @code{#} is
6485 interpreted as the beginning of a comment. Use @code{<>} instead.
6490 @chapter Examining the Symbol Table
6492 The commands described in this section allow you to inquire about the
6493 symbols (names of variables, functions and types) defined in your
6494 program. This information is inherent in the text of your program and
6495 does not change as your program executes. @value{GDBN} finds it in your
6496 program's symbol table, in the file indicated when you started @value{GDBN}
6497 (@pxref{File Options, ,Choosing files}), or by one of the
6498 file-management commands (@pxref{Files, ,Commands to specify files}).
6500 @cindex symbol names
6501 @cindex names of symbols
6502 @cindex quoting names
6503 Occasionally, you may need to refer to symbols that contain unusual
6504 characters, which @value{GDBN} ordinarily treats as word delimiters. The
6505 most frequent case is in referring to static variables in other
6506 source files (@pxref{Variables,,Program variables}). File names
6507 are recorded in object files as debugging symbols, but @value{GDBN} would
6508 ordinarily parse a typical file name, like @file{foo.c}, as the three words
6509 @samp{foo} @samp{.} @samp{c}. To allow @value{GDBN} to recognize
6510 @samp{foo.c} as a single symbol, enclose it in single quotes; for example,
6517 looks up the value of @code{x} in the scope of the file @file{foo.c}.
6520 @kindex info address
6521 @item info address @var{symbol}
6522 Describe where the data for @var{symbol} is stored. For a register
6523 variable, this says which register it is kept in. For a non-register
6524 local variable, this prints the stack-frame offset at which the variable
6527 Note the contrast with @samp{print &@var{symbol}}, which does not work
6528 at all for a register variable, and for a stack local variable prints
6529 the exact address of the current instantiation of the variable.
6532 @item whatis @var{exp}
6533 Print the data type of expression @var{exp}. @var{exp} is not
6534 actually evaluated, and any side-effecting operations (such as
6535 assignments or function calls) inside it do not take place.
6536 @xref{Expressions, ,Expressions}.
6539 Print the data type of @code{$}, the last value in the value history.
6542 @item ptype @var{typename}
6543 Print a description of data type @var{typename}. @var{typename} may be
6544 the name of a type, or for C code it may have the form
6546 @samp{class @var{class-name}},
6548 @samp{struct @var{struct-tag}}, @samp{union @var{union-tag}} or
6549 @samp{enum @var{enum-tag}}.
6551 @item ptype @var{exp}
6553 Print a description of the type of expression @var{exp}. @code{ptype}
6554 differs from @code{whatis} by printing a detailed description, instead
6555 of just the name of the type.
6557 For example, for this variable declaration:
6560 struct complex @{double real; double imag;@} v;
6564 the two commands give this output:
6568 (@value{GDBP}) whatis v
6569 type = struct complex
6570 (@value{GDBP}) ptype v
6571 type = struct complex @{
6579 As with @code{whatis}, using @code{ptype} without an argument refers to
6580 the type of @code{$}, the last value in the value history.
6583 @item info types @var{regexp}
6585 Print a brief description of all types whose name matches @var{regexp}
6586 (or all types in your program, if you supply no argument). Each
6587 complete typename is matched as though it were a complete line; thus,
6588 @samp{i type value} gives information on all types in your program whose
6589 name includes the string @code{value}, but @samp{i type ^value$} gives
6590 information only on types whose complete name is @code{value}.
6592 This command differs from @code{ptype} in two ways: first, like
6593 @code{whatis}, it does not print a detailed description; second, it
6594 lists all source files where a type is defined.
6598 Show the name of the current source file---that is, the source file for
6599 the function containing the current point of execution---and the language
6602 @kindex info sources
6604 Print the names of all source files in your program for which there is
6605 debugging information, organized into two lists: files whose symbols
6606 have already been read, and files whose symbols will be read when needed.
6608 @kindex info functions
6609 @item info functions
6610 Print the names and data types of all defined functions.
6612 @item info functions @var{regexp}
6613 Print the names and data types of all defined functions
6614 whose names contain a match for regular expression @var{regexp}.
6615 Thus, @samp{info fun step} finds all functions whose names
6616 include @code{step}; @samp{info fun ^step} finds those whose names
6617 start with @code{step}.
6619 @kindex info variables
6620 @item info variables
6621 Print the names and data types of all variables that are declared
6622 outside of functions (i.e., excluding local variables).
6624 @item info variables @var{regexp}
6625 Print the names and data types of all variables (except for local
6626 variables) whose names contain a match for regular expression
6630 This was never implemented.
6631 @kindex info methods
6633 @itemx info methods @var{regexp}
6634 The @code{info methods} command permits the user to examine all defined
6635 methods within C++ program, or (with the @var{regexp} argument) a
6636 specific set of methods found in the various C++ classes. Many
6637 C++ classes provide a large number of methods. Thus, the output
6638 from the @code{ptype} command can be overwhelming and hard to use. The
6639 @code{info-methods} command filters the methods, printing only those
6640 which match the regular-expression @var{regexp}.
6643 @cindex reloading symbols
6644 Some systems allow individual object files that make up your program to
6645 be replaced without stopping and restarting your program.
6647 For example, in VxWorks you can simply recompile a defective object file
6648 and keep on running.
6650 If you are running on one of these systems, you can allow @value{GDBN} to
6651 reload the symbols for automatically relinked modules:
6654 @kindex set symbol-reloading
6655 @item set symbol-reloading on
6656 Replace symbol definitions for the corresponding source file when an
6657 object file with a particular name is seen again.
6659 @item set symbol-reloading off
6660 Do not replace symbol definitions when re-encountering object files of
6661 the same name. This is the default state; if you are not running on a
6662 system that permits automatically relinking modules, you should leave
6663 @code{symbol-reloading} off, since otherwise @value{GDBN} may discard symbols
6664 when linking large programs, that may contain several modules (from
6665 different directories or libraries) with the same name.
6667 @kindex show symbol-reloading
6668 @item show symbol-reloading
6669 Show the current @code{on} or @code{off} setting.
6672 @kindex maint print symbols
6674 @kindex maint print psymbols
6675 @cindex partial symbol dump
6676 @item maint print symbols @var{filename}
6677 @itemx maint print psymbols @var{filename}
6678 @itemx maint print msymbols @var{filename}
6679 Write a dump of debugging symbol data into the file @var{filename}.
6680 These commands are used to debug the @value{GDBN} symbol-reading code. Only
6681 symbols with debugging data are included. If you use @samp{maint print
6682 symbols}, @value{GDBN} includes all the symbols for which it has already
6683 collected full details: that is, @var{filename} reflects symbols for
6684 only those files whose symbols @value{GDBN} has read. You can use the
6685 command @code{info sources} to find out which files these are. If you
6686 use @samp{maint print psymbols} instead, the dump shows information about
6687 symbols that @value{GDBN} only knows partially---that is, symbols defined in
6688 files that @value{GDBN} has skimmed, but not yet read completely. Finally,
6689 @samp{maint print msymbols} dumps just the minimal symbol information
6690 required for each object file from which @value{GDBN} has read some symbols.
6691 @xref{Files, ,Commands to specify files}, for a discussion of how
6692 @value{GDBN} reads symbols (in the description of @code{symbol-file}).
6696 @chapter Altering Execution
6698 Once you think you have found an error in your program, you might want to
6699 find out for certain whether correcting the apparent error would lead to
6700 correct results in the rest of the run. You can find the answer by
6701 experiment, using the @value{GDBN} features for altering execution of the
6704 For example, you can store new values into variables or memory
6707 give your program a signal, restart it
6710 restart your program
6712 at a different address, or even return prematurely from a function.
6715 * Assignment:: Assignment to variables
6716 * Jumping:: Continuing at a different address
6718 * Signaling:: Giving your program a signal
6721 * Returning:: Returning from a function
6722 * Calling:: Calling your program's functions
6723 * Patching:: Patching your program
6727 @section Assignment to variables
6730 @cindex setting variables
6731 To alter the value of a variable, evaluate an assignment expression.
6732 @xref{Expressions, ,Expressions}. For example,
6739 stores the value 4 into the variable @code{x}, and then prints the
6740 value of the assignment expression (which is 4).
6742 @xref{Languages, ,Using @value{GDBN} with Different Languages}, for more
6743 information on operators in supported languages.
6746 @kindex set variable
6747 @cindex variables, setting
6748 If you are not interested in seeing the value of the assignment, use the
6749 @code{set} command instead of the @code{print} command. @code{set} is
6750 really the same as @code{print} except that the expression's value is
6751 not printed and is not put in the value history (@pxref{Value History,
6752 ,Value history}). The expression is evaluated only for its effects.
6754 If the beginning of the argument string of the @code{set} command
6755 appears identical to a @code{set} subcommand, use the @code{set
6756 variable} command instead of just @code{set}. This command is identical
6757 to @code{set} except for its lack of subcommands. For example, if
6758 your program has a variable @code{width}, you get
6759 an error if you try to set a new value with just @samp{set width=13},
6760 because @value{GDBN} has the command @code{set width}:
6763 (@value{GDBP}) whatis width
6765 (@value{GDBP}) p width
6767 (@value{GDBP}) set width=47
6768 Invalid syntax in expression.
6772 The invalid expression, of course, is @samp{=47}. In
6773 order to actually set the program's variable @code{width}, use
6776 (@value{GDBP}) set var width=47
6779 @value{GDBN} allows more implicit conversions in assignments than C; you can
6780 freely store an integer value into a pointer variable or vice versa,
6781 and you can convert any structure to any other structure that is the
6782 same length or shorter.
6783 @comment FIXME: how do structs align/pad in these conversions?
6784 @comment /doc@cygnus.com 18dec1990
6786 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
6787 construct to generate a value of specified type at a specified address
6788 (@pxref{Expressions, ,Expressions}). For example, @code{@{int@}0x83040} refers
6789 to memory location @code{0x83040} as an integer (which implies a certain size
6790 and representation in memory), and
6793 set @{int@}0x83040 = 4
6797 stores the value 4 into that memory location.
6800 @section Continuing at a different address
6802 Ordinarily, when you continue your program, you do so at the place where
6803 it stopped, with the @code{continue} command. You can instead continue at
6804 an address of your own choosing, with the following commands:
6808 @item jump @var{linespec}
6809 Resume execution at line @var{linespec}. Execution stops again
6810 immediately if there is a breakpoint there. @xref{List, ,Printing
6811 source lines}, for a description of the different forms of
6814 The @code{jump} command does not change the current stack frame, or
6815 the stack pointer, or the contents of any memory location or any
6816 register other than the program counter. If line @var{linespec} is in
6817 a different function from the one currently executing, the results may
6818 be bizarre if the two functions expect different patterns of arguments or
6819 of local variables. For this reason, the @code{jump} command requests
6820 confirmation if the specified line is not in the function currently
6821 executing. However, even bizarre results are predictable if you are
6822 well acquainted with the machine-language code of your program.
6824 @item jump *@var{address}
6825 Resume execution at the instruction at address @var{address}.
6828 You can get much the same effect as the @code{jump} command by storing a
6829 new value into the register @code{$pc}. The difference is that this
6830 does not start your program running; it only changes the address of where it
6831 @emph{will} run when you continue. For example,
6838 makes the next @code{continue} command or stepping command execute at
6839 address @code{0x485}, rather than at the address where your program stopped.
6840 @xref{Continuing and Stepping, ,Continuing and stepping}.
6842 The most common occasion to use the @code{jump} command is to back up--
6843 perhaps with more breakpoints set--over a portion of a program that has
6844 already executed, in order to examine its execution in more detail.
6849 @section Giving your program a signal
6853 @item signal @var{signal}
6854 Resume execution where your program stopped, but immediately give it the
6855 signal @var{signal}. @var{signal} can be the name or the number of a
6856 signal. For example, on many systems @code{signal 2} and @code{signal
6857 SIGINT} are both ways of sending an interrupt signal.
6859 Alternatively, if @var{signal} is zero, continue execution without
6860 giving a signal. This is useful when your program stopped on account of
6861 a signal and would ordinary see the signal when resumed with the
6862 @code{continue} command; @samp{signal 0} causes it to resume without a
6865 @code{signal} does not repeat when you press @key{RET} a second time
6866 after executing the command.
6870 Invoking the @code{signal} command is not the same as invoking the
6871 @code{kill} utility from the shell. Sending a signal with @code{kill}
6872 causes @value{GDBN} to decide what to do with the signal depending on
6873 the signal handling tables (@pxref{Signals}). The @code{signal} command
6874 passes the signal directly to your program.
6879 @section Returning from a function
6882 @cindex returning from a function
6885 @itemx return @var{expression}
6886 You can cancel execution of a function call with the @code{return}
6887 command. If you give an
6888 @var{expression} argument, its value is used as the function's return
6892 When you use @code{return}, @value{GDBN} discards the selected stack frame
6893 (and all frames within it). You can think of this as making the
6894 discarded frame return prematurely. If you wish to specify a value to
6895 be returned, give that value as the argument to @code{return}.
6897 This pops the selected stack frame (@pxref{Selection, ,Selecting a
6898 frame}), and any other frames inside of it, leaving its caller as the
6899 innermost remaining frame. That frame becomes selected. The
6900 specified value is stored in the registers used for returning values
6903 The @code{return} command does not resume execution; it leaves the
6904 program stopped in the state that would exist if the function had just
6905 returned. In contrast, the @code{finish} command (@pxref{Continuing
6906 and Stepping, ,Continuing and stepping}) resumes execution until the
6907 selected stack frame returns naturally.
6910 @section Calling program functions
6912 @cindex calling functions
6915 @item call @var{expr}
6916 Evaluate the expression @var{expr} without displaying @code{void}
6920 You can use this variant of the @code{print} command if you want to
6921 execute a function from your program, but without cluttering the output
6922 with @code{void} returned values. If the result is not void, it
6923 is printed and saved in the value history.
6925 A new user-controlled variable, @var{call_scratch_address}, specifies
6926 the location of a scratch area to be used when @value{GDBN} calls a
6927 function in the target. This is necessary because the usual method
6928 of putting the scratch area on the stack does not work in systems that
6929 have separate instruction and data spaces.
6932 @section Patching programs
6933 @cindex patching binaries
6934 @cindex writing into executables
6936 @cindex writing into corefiles
6939 By default, @value{GDBN} opens the file containing your program's executable
6944 read-only. This prevents accidental alterations
6945 to machine code; but it also prevents you from intentionally patching
6946 your program's binary.
6948 If you'd like to be able to patch the binary, you can specify that
6949 explicitly with the @code{set write} command. For example, you might
6950 want to turn on internal debugging flags, or even to make emergency
6956 @itemx set write off
6957 If you specify @samp{set write on}, @value{GDBN} opens executable
6961 files for both reading and writing; if you specify @samp{set write
6962 off} (the default), @value{GDBN} opens them read-only.
6964 If you have already loaded a file, you must load it again (using the
6969 command) after changing @code{set write}, for your new setting to take
6974 Display whether executable files
6978 are opened for writing as well as reading.
6982 @chapter @value{GDBN} Files
6984 @value{GDBN} needs to know the file name of the program to be debugged, both in
6985 order to read its symbol table and in order to start your program.
6987 To debug a core dump of a previous run, you must also tell @value{GDBN}
6988 the name of the core dump file.
6992 * Files:: Commands to specify files
6993 * Symbol Errors:: Errors reading symbol files
6997 @section Commands to specify files
6998 @cindex symbol table
7001 @cindex core dump file
7002 You may want to specify executable and core dump file names.
7003 The usual way to do this is at start-up time, using the arguments to
7004 @value{GDBN}'s start-up commands (@pxref{Invocation, ,
7005 Getting In and Out of @value{GDBN}}).
7008 The usual way to specify an executable file name is with
7009 the command argument given when you start @value{GDBN}, (@pxref{Invocation,
7010 ,Getting In and Out of @value{GDBN}}.
7013 Occasionally it is necessary to change to a different file during a
7014 @value{GDBN} session. Or you may run @value{GDBN} and forget to specify
7015 a file you want to use. In these situations the @value{GDBN} commands
7016 to specify new files are useful.
7019 @cindex executable file
7021 @item file @var{filename}
7022 Use @var{filename} as the program to be debugged. It is read for its
7023 symbols and for the contents of pure memory. It is also the program
7024 executed when you use the @code{run} command. If you do not specify a
7025 directory and the file is not found in the @value{GDBN} working directory,
7026 @value{GDBN} uses the environment variable @code{PATH} as a list of
7027 directories to search, just as the shell does when looking for a program
7028 to run. You can change the value of this variable, for both @value{GDBN}
7029 and your program, using the @code{path} command.
7031 On systems with memory-mapped files, an auxiliary file
7032 @file{@var{filename}.syms} may hold symbol table information for
7033 @var{filename}. If so, @value{GDBN} maps in the symbol table from
7034 @file{@var{filename}.syms}, starting up more quickly. See the
7035 descriptions of the file options @samp{-mapped} and @samp{-readnow}
7036 (available on the command line, and with the commands @code{file},
7037 @code{symbol-file}, or @code{add-symbol-file}, described below),
7038 for more information.
7041 @code{file} with no argument makes @value{GDBN} discard any information it
7042 has on both executable file and the symbol table.
7045 @item exec-file @r{[} @var{filename} @r{]}
7046 Specify that the program to be run (but not the symbol table) is found
7047 in @var{filename}. @value{GDBN} searches the environment variable @code{PATH}
7048 if necessary to locate your program. Omitting @var{filename} means to
7049 discard information on the executable file.
7052 @item symbol-file @r{[} @var{filename} @r{]}
7053 Read symbol table information from file @var{filename}. @code{PATH} is
7054 searched when necessary. Use the @code{file} command to get both symbol
7055 table and program to run from the same file.
7057 @code{symbol-file} with no argument clears out @value{GDBN} information on your
7058 program's symbol table.
7060 The @code{symbol-file} command causes @value{GDBN} to forget the contents
7061 of its convenience variables, the value history, and all breakpoints and
7062 auto-display expressions. This is because they may contain pointers to
7063 the internal data recording symbols and data types, which are part of
7064 the old symbol table data being discarded inside @value{GDBN}.
7066 @code{symbol-file} does not repeat if you press @key{RET} again after
7069 When @value{GDBN} is configured for a particular environment, it
7070 understands debugging information in whatever format is the standard
7071 generated for that environment; you may use either a @sc{gnu} compiler, or
7072 other compilers that adhere to the local conventions. Best results are
7073 usually obtained from @sc{gnu} compilers; for example, using @code{@value{GCC}}
7074 you can generate debugging information for optimized code.
7076 On some kinds of object files, the @code{symbol-file} command does not
7077 normally read the symbol table in full right away. Instead, it scans
7078 the symbol table quickly to find which source files and which symbols
7079 are present. The details are read later, one source file at a time,
7082 The purpose of this two-stage reading strategy is to make @value{GDBN} start up
7083 faster. For the most part, it is invisible except for occasional
7084 pauses while the symbol table details for a particular source file are
7085 being read. (The @code{set verbose} command can turn these pauses
7086 into messages if desired. @xref{Messages/Warnings, ,Optional warnings
7089 We have not implemented the two-stage strategy for COFF yet. When the
7090 symbol table is stored in COFF format, @code{symbol-file} reads the
7091 symbol table data in full right away.
7094 @cindex reading symbols immediately
7095 @cindex symbols, reading immediately
7097 @cindex memory-mapped symbol file
7098 @cindex saving symbol table
7099 @item symbol-file @var{filename} @r{[} -readnow @r{]} @r{[} -mapped @r{]}
7100 @itemx file @var{filename} @r{[} -readnow @r{]} @r{[} -mapped @r{]}
7101 You can override the @value{GDBN} two-stage strategy for reading symbol
7102 tables by using the @samp{-readnow} option with any of the commands that
7103 load symbol table information, if you want to be sure @value{GDBN} has the
7104 entire symbol table available.
7107 If memory-mapped files are available on your system through the
7108 @code{mmap} system call, you can use another option, @samp{-mapped}, to
7109 cause @value{GDBN} to write the symbols for your program into a reusable
7110 file. Future @value{GDBN} debugging sessions map in symbol information
7111 from this auxiliary symbol file (if the program has not changed), rather
7112 than spending time reading the symbol table from the executable
7113 program. Using the @samp{-mapped} option has the same effect as
7114 starting @value{GDBN} with the @samp{-mapped} command-line option.
7116 You can use both options together, to make sure the auxiliary symbol
7117 file has all the symbol information for your program.
7119 The auxiliary symbol file for a program called @var{myprog} is called
7120 @samp{@var{myprog}.syms}. Once this file exists (so long as it is newer
7121 than the corresponding executable), @value{GDBN} always attempts to use
7122 it when you debug @var{myprog}; no special options or commands are
7125 The @file{.syms} file is specific to the host machine where you run
7126 @value{GDBN}. It holds an exact image of the internal @value{GDBN}
7127 symbol table. It cannot be shared across multiple host platforms.
7129 @c FIXME: for now no mention of directories, since this seems to be in
7130 @c flux. 13mar1992 status is that in theory GDB would look either in
7131 @c current dir or in same dir as myprog; but issues like competing
7132 @c GDB's, or clutter in system dirs, mean that in practice right now
7133 @c only current dir is used. FFish says maybe a special GDB hierarchy
7134 @c (eg rooted in val of env var GDBSYMS) could exist for mappable symbol
7139 @item core-file @r{[} @var{filename} @r{]}
7140 Specify the whereabouts of a core dump file to be used as the ``contents
7141 of memory''. Traditionally, core files contain only some parts of the
7142 address space of the process that generated them; @value{GDBN} can access the
7143 executable file itself for other parts.
7145 @code{core-file} with no argument specifies that no core file is
7148 Note that the core file is ignored when your program is actually running
7149 under @value{GDBN}. So, if you have been running your program and you wish to
7150 debug a core file instead, you must kill the subprocess in which the
7151 program is running. To do this, use the @code{kill} command
7152 (@pxref{Kill Process, ,Killing the child process}).
7155 @kindex load @var{filename}
7156 @item load @var{filename}
7158 Depending on what remote debugging facilities are configured into
7159 @value{GDBN}, the @code{load} command may be available. Where it exists, it
7160 is meant to make @var{filename} (an executable) available for debugging
7161 on the remote system---by downloading, or dynamic linking, for example.
7162 @code{load} also records the @var{filename} symbol table in @value{GDBN}, like
7163 the @code{add-symbol-file} command.
7165 If your @value{GDBN} does not have a @code{load} command, attempting to
7166 execute it gets the error message ``@code{You can't do that when your
7167 target is @dots{}}''
7170 The file is loaded at whatever address is specified in the executable.
7171 For some object file formats, you can specify the load address when you
7172 link the program; for other formats, like a.out, the object file format
7173 specifies a fixed address.
7174 @c FIXME! This would be a good place for an xref to the GNU linker doc.
7177 On VxWorks, @code{load} links @var{filename} dynamically on the
7178 current target system as well as adding its symbols in @value{GDBN}.
7182 @cindex download to Nindy-960
7183 With the Nindy interface to an Intel 960 board, @code{load}
7184 downloads @var{filename} to the 960 as well as adding its symbols in
7189 @cindex download to H8/300 or H8/500
7190 @cindex H8/300 or H8/500 download
7191 @cindex download to Hitachi SH
7192 @cindex Hitachi SH download
7193 When you select remote debugging to a Hitachi SH, H8/300, or H8/500 board
7194 (@pxref{Hitachi Remote,,@value{GDBN} and Hitachi Microprocessors}),
7195 the @code{load} command downloads your program to the Hitachi board and also
7196 opens it as the current executable target for @value{GDBN} on your host
7197 (like the @code{file} command).
7200 @code{load} does not repeat if you press @key{RET} again after using it.
7203 @kindex add-symbol-file
7204 @cindex dynamic linking
7205 @item add-symbol-file @var{filename} @var{address}
7206 @itemx add-symbol-file @var{filename} @var{address} @r{[} -readnow @r{]} @r{[} -mapped @r{]}
7207 The @code{add-symbol-file} command reads additional symbol table information
7208 from the file @var{filename}. You would use this command when @var{filename}
7209 has been dynamically loaded (by some other means) into the program that
7210 is running. @var{address} should be the memory address at which the
7211 file has been loaded; @value{GDBN} cannot figure this out for itself.
7212 You can specify @var{address} as an expression.
7214 The symbol table of the file @var{filename} is added to the symbol table
7215 originally read with the @code{symbol-file} command. You can use the
7216 @code{add-symbol-file} command any number of times; the new symbol data thus
7217 read keeps adding to the old. To discard all old symbol data instead,
7218 use the @code{symbol-file} command.
7220 @code{add-symbol-file} does not repeat if you press @key{RET} after using it.
7222 You can use the @samp{-mapped} and @samp{-readnow} options just as with
7223 the @code{symbol-file} command, to change how @value{GDBN} manages the symbol
7224 table information for @var{filename}.
7226 @kindex add-shared-symbol-file
7227 @item add-shared-symbol-file
7228 The @code{add-shared-symbol-file} command can be used only under Harris' CXUX
7229 operating system for the Motorola 88k. @value{GDBN} automatically looks for
7230 shared libraries, however if @value{GDBN} does not find yours, you can run
7231 @code{add-shared-symbol-file}. It takes no arguments.
7236 The @code{section} command changes the base address of section SECTION of
7237 the exec file to ADDR. This can be used if the exec file does not contain
7238 section addresses, (such as in the a.out format), or when the addresses
7239 specified in the file itself are wrong. Each section must be changed
7240 separately. The ``info files'' command lists all the sections and their
7247 @code{info files} and @code{info target} are synonymous; both print
7248 the current target (@pxref{Targets, ,Specifying a Debugging Target}),
7251 names of the executable and core dump files
7254 name of the executable file
7256 currently in use by @value{GDBN}, and the files from which symbols were
7257 loaded. The command @code{help target} lists all possible targets
7258 rather than current ones.
7261 All file-specifying commands allow both absolute and relative file names
7262 as arguments. @value{GDBN} always converts the file name to an absolute file
7263 name and remembers it that way.
7266 @cindex shared libraries
7267 @value{GDBN} supports SunOS, SVr4, Irix 5, and IBM RS/6000 shared libraries.
7268 @value{GDBN} automatically loads symbol definitions from shared libraries
7269 when you use the @code{run} command, or when you examine a core file.
7270 (Before you issue the @code{run} command, @value{GDBN} does not understand
7271 references to a function in a shared library, however---unless you are
7272 debugging a core file).
7273 @c FIXME: some @value{GDBN} release may permit some refs to undef
7274 @c FIXME...symbols---eg in a break cmd---assuming they are from a shared
7275 @c FIXME...lib; check this from time to time when updating manual
7278 @kindex info sharedlibrary
7281 @itemx info sharedlibrary
7282 Print the names of the shared libraries which are currently loaded.
7284 @kindex sharedlibrary
7286 @item sharedlibrary @var{regex}
7287 @itemx share @var{regex}
7289 Load shared object library symbols for files matching a
7290 Unix regular expression.
7291 As with files loaded automatically, it only loads shared libraries
7292 required by your program for a core file or after typing @code{run}. If
7293 @var{regex} is omitted all shared libraries required by your program are
7299 @section Errors reading symbol files
7301 While reading a symbol file, @value{GDBN} occasionally encounters problems,
7302 such as symbol types it does not recognize, or known bugs in compiler
7303 output. By default, @value{GDBN} does not notify you of such problems, since
7304 they are relatively common and primarily of interest to people
7305 debugging compilers. If you are interested in seeing information
7306 about ill-constructed symbol tables, you can either ask @value{GDBN} to print
7307 only one message about each such type of problem, no matter how many
7308 times the problem occurs; or you can ask @value{GDBN} to print more messages,
7309 to see how many times the problems occur, with the @code{set
7310 complaints} command (@pxref{Messages/Warnings, ,Optional warnings and
7313 The messages currently printed, and their meanings, include:
7316 @item inner block not inside outer block in @var{symbol}
7318 The symbol information shows where symbol scopes begin and end
7319 (such as at the start of a function or a block of statements). This
7320 error indicates that an inner scope block is not fully contained
7321 in its outer scope blocks.
7323 @value{GDBN} circumvents the problem by treating the inner block as if it had
7324 the same scope as the outer block. In the error message, @var{symbol}
7325 may be shown as ``@code{(don't know)}'' if the outer block is not a
7328 @item block at @var{address} out of order
7330 The symbol information for symbol scope blocks should occur in
7331 order of increasing addresses. This error indicates that it does not
7334 @value{GDBN} does not circumvent this problem, and has trouble
7335 locating symbols in the source file whose symbols it is reading. (You
7336 can often determine what source file is affected by specifying
7337 @code{set verbose on}. @xref{Messages/Warnings, ,Optional warnings and
7340 @item bad block start address patched
7342 The symbol information for a symbol scope block has a start address
7343 smaller than the address of the preceding source line. This is known
7344 to occur in the SunOS 4.1.1 (and earlier) C compiler.
7346 @value{GDBN} circumvents the problem by treating the symbol scope block as
7347 starting on the previous source line.
7349 @item bad string table offset in symbol @var{n}
7352 Symbol number @var{n} contains a pointer into the string table which is
7353 larger than the size of the string table.
7355 @value{GDBN} circumvents the problem by considering the symbol to have the
7356 name @code{foo}, which may cause other problems if many symbols end up
7359 @item unknown symbol type @code{0x@var{nn}}
7361 The symbol information contains new data types that @value{GDBN} does not yet
7362 know how to read. @code{0x@var{nn}} is the symbol type of the misunderstood
7363 information, in hexadecimal.
7365 @value{GDBN} circumvents the error by ignoring this symbol information. This
7366 usually allows you to debug your program, though certain symbols
7367 are not accessible. If you encounter such a problem and feel like
7368 debugging it, you can debug @code{@value{GDBP}} with itself, breakpoint on
7369 @code{complain}, then go up to the function @code{read_dbx_symtab} and
7370 examine @code{*bufp} to see the symbol.
7372 @item stub type has NULL name
7373 @value{GDBN} could not find the full definition for
7382 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
7384 The symbol information for a C++ member function is missing some
7385 information that recent versions of the compiler should have output
7389 @item info mismatch between compiler and debugger
7391 @value{GDBN} could not parse a type specification output by the compiler.
7395 @chapter Specifying a Debugging Target
7396 @cindex debugging target
7399 A @dfn{target} is the execution environment occupied by your program.
7401 Often, @value{GDBN} runs in the same host environment as your program; in
7402 that case, the debugging target is specified as a side effect when you
7403 use the @code{file} or @code{core} commands. When you need more
7404 flexibility---for example, running @value{GDBN} on a physically separate
7405 host, or controlling a standalone system over a serial port or a
7406 realtime system over a TCP/IP connection---you
7411 can use the @code{target} command to specify one of the target types
7412 configured for @value{GDBN} (@pxref{Target Commands, ,Commands for managing
7416 * Active Targets:: Active targets
7417 * Target Commands:: Commands for managing targets
7418 * Remote:: Remote debugging
7421 @node Active Targets
7422 @section Active targets
7423 @cindex stacking targets
7424 @cindex active targets
7425 @cindex multiple targets
7428 There are three classes of targets: processes, core files, and
7429 executable files. @value{GDBN} can work concurrently on up to three active
7430 targets, one in each class. This allows you to (for example) start a
7431 process and inspect its activity without abandoning your work on a core
7434 For example, if you execute @samp{gdb a.out}, then the executable file
7435 @code{a.out} is the only active target. If you designate a core file as
7436 well---presumably from a prior run that crashed and coredumped---then
7437 @value{GDBN} has two active targets and uses them in tandem, looking
7438 first in the corefile target, then in the executable file, to satisfy
7439 requests for memory addresses. (Typically, these two classes of target
7440 are complementary, since core files contain only a program's
7441 read-write memory---variables and so on---plus machine status, while
7442 executable files contain only the program text and initialized data.)
7445 When you type @code{run}, your executable file becomes an active process
7446 target as well. When a process target is active, all @value{GDBN} commands
7447 requesting memory addresses refer to that target; addresses in an
7451 executable file target are obscured while the process
7455 Use the @code{exec-file} command to select a
7456 new executable target (@pxref{Files, ,Commands to specify
7460 Use the @code{core-file} and @code{exec-file} commands to select a
7461 new core file or executable target (@pxref{Files, ,Commands to specify
7462 files}). To specify as a target a process that is already running, use
7463 the @code{attach} command (@pxref{Attach, ,Debugging an
7464 already-running process}).
7467 @node Target Commands
7468 @section Commands for managing targets
7471 @item target @var{type} @var{parameters}
7472 Connects the @value{GDBN} host environment to a target
7477 machine or process. A target is typically a protocol for talking to
7478 debugging facilities. You use the argument @var{type} to specify the
7479 type or protocol of the target machine.
7481 Further @var{parameters} are interpreted by the target protocol, but
7482 typically include things like device names or host names to connect
7483 with, process numbers, and baud rates.
7486 The @code{target} command does not repeat if you press @key{RET} again
7487 after executing the command.
7491 Displays the names of all targets available. To display targets
7492 currently selected, use either @code{info target} or @code{info files}
7493 (@pxref{Files, ,Commands to specify files}).
7495 @item help target @var{name}
7496 Describe a particular target, including any parameters necessary to
7499 @kindex set gnutarget
7500 @item set gnutarget @var{args}
7501 @value{GDBN}uses its own library BFD to read your files. @value{GDBN}
7502 knows whether it is reading an @dfn{executable},
7503 a @dfn{core}, or a @dfn{.o} file, however you can specify the file format
7504 with the @code{set gnutarget} command. Unlike most @code{target} commands,
7505 with @code{gnutarget} the @code{target} refers to a program, not a machine.
7507 @emph{Warning:} To specify a file format with @code{set gnutarget},
7508 you must know the actual BFD name.
7510 @noindent @xref{Files, , Commands to specify files}.
7512 @kindex show gnutarget
7513 @item show gnutarget
7514 Use the @code{show gnutarget} command to display what file format
7515 @code{gnutarget} is set to read. If you have not set @code{gnutarget},
7516 @value{GDBN} will determine the file format for each file automatically
7517 and @code{show gnutarget} displays @code{The current BDF target is "auto"}.
7520 Here are some common targets (available, or not, depending on the GDB
7525 @item target exec @var{program}
7526 An executable file. @samp{target exec @var{program}} is the same as
7527 @samp{exec-file @var{program}}.
7531 @item target core @var{filename}
7532 A core dump file. @samp{target core @var{filename}} is the same as
7533 @samp{core-file @var{filename}}.
7537 @kindex target remote
7538 @item target remote @var{dev}
7539 Remote serial target in GDB-specific protocol. The argument @var{dev}
7540 specifies what serial device to use for the connection (e.g.
7541 @file{/dev/ttya}). @xref{Remote, ,Remote debugging}. @code{target remote}
7542 now supports the @code{load} command. This is only useful if you have
7543 some other way of getting the stub to the target system, and you can put
7544 it somewhere in memory where it won't get clobbered by the download.
7550 CPU simulator. @xref{Simulator,,Simulated CPU Target}.
7555 @item target udi @var{keyword}
7556 Remote AMD29K target, using the AMD UDI protocol. The @var{keyword}
7557 argument specifies which 29K board or simulator to use. @xref{UDI29K
7558 Remote,,The UDI protocol for AMD29K}.
7560 @kindex target amd-eb
7561 @item target amd-eb @var{dev} @var{speed} @var{PROG}
7563 Remote PC-resident AMD EB29K board, attached over serial lines.
7564 @var{dev} is the serial device, as for @code{target remote};
7565 @var{speed} allows you to specify the linespeed; and @var{PROG} is the
7566 name of the program to be debugged, as it appears to DOS on the PC.
7567 @xref{EB29K Remote, ,The EBMON protocol for AMD29K}.
7572 @item target hms @var{dev}
7573 A Hitachi SH, H8/300, or H8/500 board, attached via serial line to your host.
7574 @ifclear H8EXCLUSIVE
7575 Use special commands @code{device} and @code{speed} to control the serial
7576 line and the communications speed used.
7578 @xref{Hitachi Remote,,@value{GDBN} and Hitachi Microprocessors}.
7582 @kindex target nindy
7583 @item target nindy @var{devicename}
7584 An Intel 960 board controlled by a Nindy Monitor. @var{devicename} is
7585 the name of the serial device to use for the connection, e.g.
7586 @file{/dev/ttya}. @xref{i960-Nindy Remote, ,@value{GDBN} with a remote i960 (Nindy)}.
7590 @kindex target st2000
7591 @item target st2000 @var{dev} @var{speed}
7592 A Tandem ST2000 phone switch, running Tandem's STDBUG protocol. @var{dev}
7593 is the name of the device attached to the ST2000 serial line;
7594 @var{speed} is the communication line speed. The arguments are not used
7595 if @value{GDBN} is configured to connect to the ST2000 using TCP or Telnet.
7596 @xref{ST2000 Remote,,@value{GDBN} with a Tandem ST2000}.
7600 @kindex target vxworks
7601 @item target vxworks @var{machinename}
7602 A VxWorks system, attached via TCP/IP. The argument @var{machinename}
7603 is the target system's machine name or IP address.
7604 @xref{VxWorks Remote, ,@value{GDBN} and VxWorks}.
7608 @item target bug @var{dev}
7609 BUG monitor, running on a MVME187 (m88k) board.
7611 @kindex target cpu32bug
7612 @item target cpu32bug @var{dev}
7613 CPU32BUG monitor, running on a CPU32 (M68K) board.
7615 @kindex target op50n
7616 @item target op50n @var{dev}
7617 OP50N monitor, running on an OKI HPPA board.
7620 @item target w89k @var{dev}
7621 W89K monitor, running on a Winbond HPPA board.
7624 @item target est @var{dev}
7625 EST-300 ICE monitor, running on a CPU32 (M68K) board.
7627 @kindex target rom68k
7628 @item target rom68k @var{dev}
7629 ROM 68K monitor, running on an IDP board.
7631 @kindex target array
7632 @item target array @var{dev}
7633 Array Tech LSI33K RAID controller board.
7635 @kindex target sparclite
7636 @item target sparclite @var{dev}
7637 Fujitsu sparclite boards, used only for the purpose of loading.
7638 You must use an additional command to debug the program.
7639 For example: target remote @var{dev} using @value{GDBN} standard
7644 Different targets are available on different configurations of @value{GDBN};
7645 your configuration may have more or fewer targets.
7648 @section Choosing target byte order
7649 @cindex choosing target byte order
7650 @cindex target byte order
7651 @kindex set endian big
7652 @kindex set endian little
7653 @kindex set endian auto
7656 You can now choose which byte order to use with a target system.
7657 Use the @code{set endian big} and @code{set endian little} commands.
7658 Use the @code{set endian auto} command to instruct
7659 @value{GDBN} to use the byte order associated with the executable.
7660 You can see the current setting for byte order with the @code{show endian}
7663 @emph{Warning:} Currently, only embedded MIPS configurations support
7664 dynamic selection of target byte order.
7667 @section Remote debugging
7668 @cindex remote debugging
7670 If you are trying to debug a program running on a machine that cannot run
7671 @value{GDBN} in the usual way, it is often useful to use remote debugging.
7672 For example, you might use remote debugging on an operating system kernel,
7673 or on a small system which does not have a general purpose operating system
7674 powerful enough to run a full-featured debugger.
7676 Some configurations of @value{GDBN} have special serial or TCP/IP interfaces
7677 to make this work with particular debugging targets. In addition,
7678 @value{GDBN} comes with a generic serial protocol (specific to @value{GDBN},
7679 but not specific to any particular target system) which you can use if you
7680 write the remote stubs---the code that runs on the remote system to
7681 communicate with @value{GDBN}.
7683 Other remote targets may be available in your
7684 configuration of @value{GDBN}; use @code{help target} to list them.
7687 @c Text on starting up GDB in various specific cases; it goes up front
7688 @c in manuals configured for any of those particular situations, here
7692 * Remote Serial:: @value{GDBN} remote serial protocol
7695 * i960-Nindy Remote:: @value{GDBN} with a remote i960 (Nindy)
7698 * UDI29K Remote:: The UDI protocol for AMD29K
7699 * EB29K Remote:: The EBMON protocol for AMD29K
7702 * VxWorks Remote:: @value{GDBN} and VxWorks
7705 * ST2000 Remote:: @value{GDBN} with a Tandem ST2000
7708 * Hitachi Remote:: @value{GDBN} and Hitachi Microprocessors
7711 * MIPS Remote:: @value{GDBN} and MIPS boards
7714 * Sparclet Remote:: @value{GDBN} and Sparclet boards
7717 * Simulator:: Simulated CPU target
7721 @include remote.texi
7724 @node Controlling GDB
7725 @chapter Controlling @value{GDBN}
7727 You can alter the way @value{GDBN} interacts with you by using
7728 the @code{set} command. For commands controlling how @value{GDBN} displays
7729 data, @pxref{Print Settings, ,Print settings}; other settings are described
7734 * Editing:: Command editing
7735 * History:: Command history
7736 * Screen Size:: Screen size
7738 * Messages/Warnings:: Optional warnings and messages
7746 @value{GDBN} indicates its readiness to read a command by printing a string
7747 called the @dfn{prompt}. This string is normally @samp{(@value{GDBP})}. You
7748 can change the prompt string with the @code{set prompt} command. For
7749 instance, when debugging @value{GDBN} with @value{GDBN}, it is useful to change
7750 the prompt in one of the @value{GDBN} sessions so that you can always tell
7751 which one you are talking to.
7753 @emph{Note:} @code{set prompt} no longer adds a space for you after the
7754 prompt you set. This allows you to set a prompt which ends in a space
7755 or a prompt that does not.
7759 @item set prompt @var{newprompt}
7760 Directs @value{GDBN} to use @var{newprompt} as its prompt string henceforth.
7764 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
7768 @section Command editing
7770 @cindex command line editing
7772 @value{GDBN} reads its input commands via the @dfn{readline} interface. This
7773 @sc{gnu} library provides consistent behavior for programs which provide a
7774 command line interface to the user. Advantages are @sc{gnu} Emacs-style
7775 or @dfn{vi}-style inline editing of commands, @code{csh}-like history
7776 substitution, and a storage and recall of command history across
7779 You may control the behavior of command line editing in @value{GDBN} with the
7786 @itemx set editing on
7787 Enable command line editing (enabled by default).
7789 @item set editing off
7790 Disable command line editing.
7792 @kindex show editing
7794 Show whether command line editing is enabled.
7798 @section Command history
7800 @value{GDBN} can keep track of the commands you type during your
7801 debugging sessions, so that you can be certain of precisely what
7802 happened. Use these commands to manage the @value{GDBN} command
7806 @cindex history substitution
7807 @cindex history file
7808 @kindex set history filename
7810 @item set history filename @var{fname}
7811 Set the name of the @value{GDBN} command history file to @var{fname}.
7812 This is the file where @value{GDBN} reads an initial command history
7813 list, and where it writes the command history from this session when it
7814 exits. You can access this list through history expansion or through
7815 the history command editing characters listed below. This file defaults
7816 to the value of the environment variable @code{GDBHISTFILE}, or to
7817 @file{./.gdb_history} if this variable is not set.
7819 @cindex history save
7820 @kindex set history save
7821 @item set history save
7822 @itemx set history save on
7823 Record command history in a file, whose name may be specified with the
7824 @code{set history filename} command. By default, this option is disabled.
7826 @item set history save off
7827 Stop recording command history in a file.
7829 @cindex history size
7830 @kindex set history size
7831 @item set history size @var{size}
7832 Set the number of commands which @value{GDBN} keeps in its history list.
7833 This defaults to the value of the environment variable
7834 @code{HISTSIZE}, or to 256 if this variable is not set.
7837 @cindex history expansion
7838 History expansion assigns special meaning to the character @kbd{!}.
7839 @ifset have-readline-appendices
7840 @xref{Event Designators}.
7843 Since @kbd{!} is also the logical not operator in C, history expansion
7844 is off by default. If you decide to enable history expansion with the
7845 @code{set history expansion on} command, you may sometimes need to
7846 follow @kbd{!} (when it is used as logical not, in an expression) with
7847 a space or a tab to prevent it from being expanded. The readline
7848 history facilities do not attempt substitution on the strings
7849 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
7851 The commands to control history expansion are:
7854 @kindex set history expansion
7855 @item set history expansion on
7856 @itemx set history expansion
7857 Enable history expansion. History expansion is off by default.
7859 @item set history expansion off
7860 Disable history expansion.
7862 The readline code comes with more complete documentation of
7863 editing and history expansion features. Users unfamiliar with @sc{gnu} Emacs
7864 or @code{vi} may wish to read it.
7865 @ifset have-readline-appendices
7866 @xref{Command Line Editing}.
7870 @kindex show history
7872 @itemx show history filename
7873 @itemx show history save
7874 @itemx show history size
7875 @itemx show history expansion
7876 These commands display the state of the @value{GDBN} history parameters.
7877 @code{show history} by itself displays all four states.
7882 @kindex show commands
7884 Display the last ten commands in the command history.
7886 @item show commands @var{n}
7887 Print ten commands centered on command number @var{n}.
7889 @item show commands +
7890 Print ten commands just after the commands last printed.
7894 @section Screen size
7895 @cindex size of screen
7896 @cindex pauses in output
7898 Certain commands to @value{GDBN} may produce large amounts of
7899 information output to the screen. To help you read all of it,
7900 @value{GDBN} pauses and asks you for input at the end of each page of
7901 output. Type @key{RET} when you want to continue the output, or @kbd{q}
7902 to discard the remaining output. Also, the screen width setting
7903 determines when to wrap lines of output. Depending on what is being
7904 printed, @value{GDBN} tries to break the line at a readable place,
7905 rather than simply letting it overflow onto the following line.
7907 Normally @value{GDBN} knows the size of the screen from the termcap data base
7908 together with the value of the @code{TERM} environment variable and the
7909 @code{stty rows} and @code{stty cols} settings. If this is not correct,
7910 you can override it with the @code{set height} and @code{set
7918 @item set height @var{lpp}
7920 @itemx set width @var{cpl}
7922 These @code{set} commands specify a screen height of @var{lpp} lines and
7923 a screen width of @var{cpl} characters. The associated @code{show}
7924 commands display the current settings.
7926 If you specify a height of zero lines, @value{GDBN} does not pause during
7927 output no matter how long the output is. This is useful if output is to a
7928 file or to an editor buffer.
7930 Likewise, you can specify @samp{set width 0} to prevent @value{GDBN}
7931 from wrapping its output.
7936 @cindex number representation
7937 @cindex entering numbers
7939 You can always enter numbers in octal, decimal, or hexadecimal in @value{GDBN} by
7940 the usual conventions: octal numbers begin with @samp{0}, decimal
7941 numbers end with @samp{.}, and hexadecimal numbers begin with @samp{0x}.
7942 Numbers that begin with none of these are, by default, entered in base
7943 10; likewise, the default display for numbers---when no particular
7944 format is specified---is base 10. You can change the default base for
7945 both input and output with the @code{set radix} command.
7948 @kindex set input-radix
7949 @item set input-radix @var{base}
7950 Set the default base for numeric input. Supported choices
7951 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
7952 specified either unambiguously or using the current default radix; for
7962 sets the base to decimal. On the other hand, @samp{set radix 10}
7963 leaves the radix unchanged no matter what it was.
7965 @kindex set output-radix
7966 @item set output-radix @var{base}
7967 Set the default base for numeric display. Supported choices
7968 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
7969 specified either unambiguously or using the current default radix.
7971 @kindex show input-radix
7972 @item show input-radix
7973 Display the current default base for numeric input.
7975 @kindex show output-radix
7976 @item show output-radix
7977 Display the current default base for numeric display.
7980 @node Messages/Warnings
7981 @section Optional warnings and messages
7983 By default, @value{GDBN} is silent about its inner workings. If you are running
7984 on a slow machine, you may want to use the @code{set verbose} command.
7985 This makes @value{GDBN} tell you when it does a lengthy internal operation, so
7986 you will not think it has crashed.
7988 Currently, the messages controlled by @code{set verbose} are those
7989 which announce that the symbol table for a source file is being read;
7990 see @code{symbol-file} in @ref{Files, ,Commands to specify files}.
7994 @item set verbose on
7995 Enables @value{GDBN} output of certain informational messages.
7997 @item set verbose off
7998 Disables @value{GDBN} output of certain informational messages.
8000 @kindex show verbose
8002 Displays whether @code{set verbose} is on or off.
8005 By default, if @value{GDBN} encounters bugs in the symbol table of an object
8006 file, it is silent; but if you are debugging a compiler, you may find
8007 this information useful (@pxref{Symbol Errors, ,Errors reading symbol files}).
8010 @kindex set complaints
8011 @item set complaints @var{limit}
8012 Permits @value{GDBN} to output @var{limit} complaints about each type of unusual
8013 symbols before becoming silent about the problem. Set @var{limit} to
8014 zero to suppress all complaints; set it to a large number to prevent
8015 complaints from being suppressed.
8017 @kindex show complaints
8018 @item show complaints
8019 Displays how many symbol complaints @value{GDBN} is permitted to produce.
8022 By default, @value{GDBN} is cautious, and asks what sometimes seems to be a
8023 lot of stupid questions to confirm certain commands. For example, if
8024 you try to run a program which is already running:
8028 The program being debugged has been started already.
8029 Start it from the beginning? (y or n)
8032 If you are willing to unflinchingly face the consequences of your own
8033 commands, you can disable this ``feature'':
8038 @cindex confirmation
8039 @cindex stupid questions
8040 @item set confirm off
8041 Disables confirmation requests.
8043 @item set confirm on
8044 Enables confirmation requests (the default).
8046 @kindex show confirm
8048 Displays state of confirmation requests.
8052 @chapter Canned Sequences of Commands
8054 Aside from breakpoint commands (@pxref{Break Commands, ,Breakpoint
8055 command lists}), @value{GDBN} provides two ways to store sequences of commands
8056 for execution as a unit: user-defined commands and command files.
8059 * Define:: User-defined commands
8060 * Hooks:: User-defined command hooks
8061 * Command Files:: Command files
8062 * Output:: Commands for controlled output
8066 @section User-defined commands
8068 @cindex user-defined command
8069 A @dfn{user-defined command} is a sequence of @value{GDBN} commands to which
8070 you assign a new name as a command. This is done with the @code{define}
8071 command. User commands may accept up to 10 arguments separated by whitespace.
8072 Arguments are accessed within the user command via @var{$arg0@dots{}$arg9}.
8077 print $arg0 + $arg1 + $arg2
8080 @noindent To execute the command use:
8086 @noindent This defines the command @code{adder}, which prints the sum of
8087 its three arguments. Note the arguments are text substitutions, so they may
8088 reference variables, use complex expressions, or even perform inferior
8093 @item define @var{commandname}
8094 Define a command named @var{commandname}. If there is already a command
8095 by that name, you are asked to confirm that you want to redefine it.
8097 The definition of the command is made up of other @value{GDBN} command lines,
8098 which are given following the @code{define} command. The end of these
8099 commands is marked by a line containing @code{end}.
8104 Takes a single argument, which is an expression to evaluate.
8105 It is followed by a series of commands that are executed
8106 only if the expression is true (nonzero).
8107 There can then optionally be a line @code{else}, followed
8108 by a series of commands that are only executed if the expression
8109 was false. The end of the list is marked by a line containing @code{end}.
8113 The syntax is similar to @code{if}: the command takes a single argument,
8114 which is an expression to evaluate, and must be followed by the commands to
8115 execute, one per line, terminated by an @code{end}.
8116 The commands are executed repeatedly as long as the expression
8120 @item document @var{commandname}
8121 Document the user-defined command @var{commandname}, so that it can be
8122 accessed by @code{help}. The command @var{commandname} must already be
8123 defined. This command reads lines of documentation just as @code{define}
8124 reads the lines of the command definition, ending with @code{end}.
8125 After the @code{document} command is finished, @code{help} on command
8126 @var{commandname} displays the documentation you have written.
8128 You may use the @code{document} command again to change the
8129 documentation of a command. Redefining the command with @code{define}
8130 does not change the documentation.
8132 @kindex help user-defined
8133 @item help user-defined
8134 List all user-defined commands, with the first line of the documentation
8139 @itemx show user @var{commandname}
8140 Display the @value{GDBN} commands used to define @var{commandname} (but not its
8141 documentation). If no @var{commandname} is given, display the
8142 definitions for all user-defined commands.
8145 When user-defined commands are executed, the
8146 commands of the definition are not printed. An error in any command
8147 stops execution of the user-defined command.
8149 If used interactively, commands that would ask for confirmation proceed
8150 without asking when used inside a user-defined command. Many @value{GDBN}
8151 commands that normally print messages to say what they are doing omit the
8152 messages when used in a user-defined command.
8155 @section User-defined command hooks
8156 @cindex command files
8158 You may define @emph{hooks}, which are a special kind of user-defined
8159 command. Whenever you run the command @samp{foo}, if the user-defined
8160 command @samp{hook-foo} exists, it is executed (with no arguments)
8161 before that command.
8163 In addition, a pseudo-command, @samp{stop} exists. Defining
8164 (@samp{hook-stop}) makes the associated commands execute every time
8165 execution stops in your program: before breakpoint commands are run,
8166 displays are printed, or the stack frame is printed.
8169 For example, to ignore @code{SIGALRM} signals while
8170 single-stepping, but treat them normally during normal execution,
8175 handle SIGALRM nopass
8182 define hook-continue
8188 You can define a hook for any single-word command in @value{GDBN}, but
8189 not for command aliases; you should define a hook for the basic command
8190 name, e.g. @code{backtrace} rather than @code{bt}.
8191 @c FIXME! So how does Joe User discover whether a command is an alias
8193 If an error occurs during the execution of your hook, execution of
8194 @value{GDBN} commands stops and @value{GDBN} issues a prompt
8195 (before the command that you actually typed had a chance to run).
8197 If you try to define a hook which does not match any known command, you
8198 get a warning from the @code{define} command.
8201 @section Command files
8203 @cindex command files
8204 A command file for @value{GDBN} is a file of lines that are @value{GDBN}
8205 commands. Comments (lines starting with @kbd{#}) may also be included.
8206 An empty line in a command file does nothing; it does not mean to repeat
8207 the last command, as it would from the terminal.
8210 @cindex @file{@value{GDBINIT}}
8211 When you start @value{GDBN}, it automatically executes commands from its
8212 @dfn{init files}. These are files named @file{@value{GDBINIT}}.
8213 @value{GDBN} reads the init file (if any) in your home directory, then
8214 processes command line options and operands, and then reads the init
8215 file (if any) in the current working directory. This is so the init
8216 file in your home directory can set options (such as @code{set
8217 complaints}) which affect the processing of the command line options and
8218 operands. The init files are not executed if you use the @samp{-nx}
8219 option; @pxref{Mode Options, ,Choosing modes}.
8222 @cindex init file name
8223 On some configurations of @value{GDBN}, the init file is known by a
8224 different name (these are typically environments where a specialized
8225 form of @value{GDBN} may need to coexist with other forms,
8226 hence a different name
8227 for the specialized version's init file). These are the environments
8228 with special init file names:
8233 VxWorks (Wind River Systems real-time OS): @samp{.vxgdbinit}
8235 @kindex .os68gdbinit
8237 OS68K (Enea Data Systems real-time OS): @samp{.os68gdbinit}
8241 ES-1800 (Ericsson Telecom AB M68000 emulator): @samp{.esgdbinit}
8245 You can also request the execution of a command file with the
8246 @code{source} command:
8250 @item source @var{filename}
8251 Execute the command file @var{filename}.
8254 The lines in a command file are executed sequentially. They are not
8255 printed as they are executed. An error in any command terminates execution
8256 of the command file.
8258 Commands that would ask for confirmation if used interactively proceed
8259 without asking when used in a command file. Many @value{GDBN} commands that
8260 normally print messages to say what they are doing omit the messages
8261 when called from command files.
8264 @section Commands for controlled output
8266 During the execution of a command file or a user-defined command, normal
8267 @value{GDBN} output is suppressed; the only output that appears is what is
8268 explicitly printed by the commands in the definition. This section
8269 describes three commands useful for generating exactly the output you
8274 @item echo @var{text}
8275 @c I do not consider backslash-space a standard C escape sequence
8276 @c because it is not in ANSI.
8277 Print @var{text}. Nonprinting characters can be included in
8278 @var{text} using C escape sequences, such as @samp{\n} to print a
8279 newline. @strong{No newline is printed unless you specify one.}
8280 In addition to the standard C escape sequences, a backslash followed
8281 by a space stands for a space. This is useful for displaying a
8282 string with spaces at the beginning or the end, since leading and
8283 trailing spaces are otherwise trimmed from all arguments.
8284 To print @samp{@w{ }and foo =@w{ }}, use the command
8285 @samp{echo \@w{ }and foo = \@w{ }}.
8287 A backslash at the end of @var{text} can be used, as in C, to continue
8288 the command onto subsequent lines. For example,
8291 echo This is some text\n\
8292 which is continued\n\
8293 onto several lines.\n
8296 produces the same output as
8299 echo This is some text\n
8300 echo which is continued\n
8301 echo onto several lines.\n
8305 @item output @var{expression}
8306 Print the value of @var{expression} and nothing but that value: no
8307 newlines, no @samp{$@var{nn} = }. The value is not entered in the
8308 value history either. @xref{Expressions, ,Expressions}, for more information
8311 @item output/@var{fmt} @var{expression}
8312 Print the value of @var{expression} in format @var{fmt}. You can use
8313 the same formats as for @code{print}. @xref{Output Formats,,Output
8314 formats}, for more information.
8317 @item printf @var{string}, @var{expressions}@dots{}
8318 Print the values of the @var{expressions} under the control of
8319 @var{string}. The @var{expressions} are separated by commas and may be
8320 either numbers or pointers. Their values are printed as specified by
8321 @var{string}, exactly as if your program were to execute the C
8325 printf (@var{string}, @var{expressions}@dots{});
8328 For example, you can print two values in hex like this:
8331 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
8334 The only backslash-escape sequences that you can use in the format
8335 string are the simple ones that consist of backslash followed by a
8341 @chapter Using @value{GDBN} under @sc{gnu} Emacs
8344 @cindex @sc{gnu} Emacs
8345 A special interface allows you to use @sc{gnu} Emacs to view (and
8346 edit) the source files for the program you are debugging with
8349 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
8350 executable file you want to debug as an argument. This command starts
8351 @value{GDBN} as a subprocess of Emacs, with input and output through a newly
8352 created Emacs buffer.
8354 Using @value{GDBN} under Emacs is just like using @value{GDBN} normally except for two
8359 All ``terminal'' input and output goes through the Emacs buffer.
8362 This applies both to @value{GDBN} commands and their output, and to the input
8363 and output done by the program you are debugging.
8365 This is useful because it means that you can copy the text of previous
8366 commands and input them again; you can even use parts of the output
8369 All the facilities of Emacs' Shell mode are available for interacting
8370 with your program. In particular, you can send signals the usual
8371 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
8376 @value{GDBN} displays source code through Emacs.
8379 Each time @value{GDBN} displays a stack frame, Emacs automatically finds the
8380 source file for that frame and puts an arrow (@samp{=>}) at the
8381 left margin of the current line. Emacs uses a separate buffer for
8382 source display, and splits the screen to show both your @value{GDBN} session
8385 Explicit @value{GDBN} @code{list} or search commands still produce output as
8386 usual, but you probably have no reason to use them from Emacs.
8389 @emph{Warning:} If the directory where your program resides is not your
8390 current directory, it can be easy to confuse Emacs about the location of
8391 the source files, in which case the auxiliary display buffer does not
8392 appear to show your source. @value{GDBN} can find programs by searching your
8393 environment's @code{PATH} variable, so the @value{GDBN} input and output
8394 session proceeds normally; but Emacs does not get enough information
8395 back from @value{GDBN} to locate the source files in this situation. To
8396 avoid this problem, either start @value{GDBN} mode from the directory where
8397 your program resides, or specify an absolute file name when prompted for the
8398 @kbd{M-x gdb} argument.
8400 A similar confusion can result if you use the @value{GDBN} @code{file} command to
8401 switch to debugging a program in some other location, from an existing
8402 @value{GDBN} buffer in Emacs.
8405 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If
8406 you need to call @value{GDBN} by a different name (for example, if you keep
8407 several configurations around, with different names) you can set the
8408 Emacs variable @code{gdb-command-name}; for example,
8411 (setq gdb-command-name "mygdb")
8415 (preceded by @kbd{ESC ESC}, or typed in the @code{*scratch*} buffer, or
8416 in your @file{.emacs} file) makes Emacs call the program named
8417 ``@code{mygdb}'' instead.
8419 In the @value{GDBN} I/O buffer, you can use these special Emacs commands in
8420 addition to the standard Shell mode commands:
8424 Describe the features of Emacs' @value{GDBN} Mode.
8427 Execute to another source line, like the @value{GDBN} @code{step} command; also
8428 update the display window to show the current file and location.
8431 Execute to next source line in this function, skipping all function
8432 calls, like the @value{GDBN} @code{next} command. Then update the display window
8433 to show the current file and location.
8436 Execute one instruction, like the @value{GDBN} @code{stepi} command; update
8437 display window accordingly.
8440 Execute to next instruction, using the @value{GDBN} @code{nexti} command; update
8441 display window accordingly.
8444 Execute until exit from the selected stack frame, like the @value{GDBN}
8445 @code{finish} command.
8448 Continue execution of your program, like the @value{GDBN} @code{continue}
8451 @emph{Warning:} In Emacs v19, this command is @kbd{C-c C-p}.
8454 Go up the number of frames indicated by the numeric argument
8455 (@pxref{Arguments, , Numeric Arguments, Emacs, The @sc{gnu} Emacs Manual}),
8456 like the @value{GDBN} @code{up} command.
8458 @emph{Warning:} In Emacs v19, this command is @kbd{C-c C-u}.
8461 Go down the number of frames indicated by the numeric argument, like the
8462 @value{GDBN} @code{down} command.
8464 @emph{Warning:} In Emacs v19, this command is @kbd{C-c C-d}.
8467 Read the number where the cursor is positioned, and insert it at the end
8468 of the @value{GDBN} I/O buffer. For example, if you wish to disassemble code
8469 around an address that was displayed earlier, type @kbd{disassemble};
8470 then move the cursor to the address display, and pick up the
8471 argument for @code{disassemble} by typing @kbd{C-x &}.
8473 You can customize this further by defining elements of the list
8474 @code{gdb-print-command}; once it is defined, you can format or
8475 otherwise process numbers picked up by @kbd{C-x &} before they are
8476 inserted. A numeric argument to @kbd{C-x &} indicates that you
8477 wish special formatting, and also acts as an index to pick an element of the
8478 list. If the list element is a string, the number to be inserted is
8479 formatted using the Emacs function @code{format}; otherwise the number
8480 is passed as an argument to the corresponding list element.
8483 In any source file, the Emacs command @kbd{C-x SPC} (@code{gdb-break})
8484 tells @value{GDBN} to set a breakpoint on the source line point is on.
8486 If you accidentally delete the source-display buffer, an easy way to get
8487 it back is to type the command @code{f} in the @value{GDBN} buffer, to
8488 request a frame display; when you run under Emacs, this recreates
8489 the source buffer if necessary to show you the context of the current
8492 The source files displayed in Emacs are in ordinary Emacs buffers
8493 which are visiting the source files in the usual way. You can edit
8494 the files with these buffers if you wish; but keep in mind that @value{GDBN}
8495 communicates with Emacs in terms of line numbers. If you add or
8496 delete lines from the text, the line numbers that @value{GDBN} knows cease
8497 to correspond properly with the code.
8499 @c The following dropped because Epoch is nonstandard. Reactivate
8500 @c if/when v19 does something similar. ---doc@cygnus.com 19dec1990
8502 @kindex Emacs Epoch environment
8506 Version 18 of @sc{gnu} Emacs has a built-in window system
8507 called the @code{epoch}
8508 environment. Users of this environment can use a new command,
8509 @code{inspect} which performs identically to @code{print} except that
8510 each value is printed in its own window.
8516 @chapter Using @value{GDBN} with Energize
8519 The Energize Programming System is an integrated development environment
8520 that includes a point-and-click interface to many programming tools.
8521 When you use @value{GDBN} in this environment, you can use the standard
8522 Energize graphical interface to drive @value{GDBN}; you can also, if you
8523 choose, type @value{GDBN} commands as usual in a debugging window. Even if
8524 you use the graphical interface, the debugging window (which uses Emacs,
8525 and resembles the standard @sc{gnu} Emacs interface to
8526 @value{GDBN}) displays the
8527 equivalent commands, so that the history of your debugging session is
8530 When Energize starts up a @value{GDBN} session, it uses one of the
8531 command-line options @samp{-energize} or @samp{-cadillac} (``cadillac''
8532 is the name of the communications protocol used by the Energize system).
8533 This option makes @value{GDBN} run as one of the tools in the Energize Tool
8534 Set: it sends all output to the Energize kernel, and accept input from
8537 See the user manual for the Energize Programming System for
8538 information on how to use the Energize graphical interface and the other
8539 development tools that Energize integrates with @value{GDBN}.
8544 @chapter Reporting Bugs in @value{GDBN}
8545 @cindex bugs in @value{GDBN}
8546 @cindex reporting bugs in @value{GDBN}
8548 Your bug reports play an essential role in making @value{GDBN} reliable.
8550 Reporting a bug may help you by bringing a solution to your problem, or it
8551 may not. But in any case the principal function of a bug report is to help
8552 the entire community by making the next version of @value{GDBN} work better. Bug
8553 reports are your contribution to the maintenance of @value{GDBN}.
8555 In order for a bug report to serve its purpose, you must include the
8556 information that enables us to fix the bug.
8559 * Bug Criteria:: Have you found a bug?
8560 * Bug Reporting:: How to report bugs
8564 @section Have you found a bug?
8565 @cindex bug criteria
8567 If you are not sure whether you have found a bug, here are some guidelines:
8570 @cindex fatal signal
8571 @cindex debugger crash
8572 @cindex crash of debugger
8574 If the debugger gets a fatal signal, for any input whatever, that is a
8575 @value{GDBN} bug. Reliable debuggers never crash.
8577 @cindex error on valid input
8579 If @value{GDBN} produces an error message for valid input, that is a bug.
8581 @cindex invalid input
8583 If @value{GDBN} does not produce an error message for invalid input,
8584 that is a bug. However, you should note that your idea of
8585 ``invalid input'' might be our idea of ``an extension'' or ``support
8586 for traditional practice''.
8589 If you are an experienced user of debugging tools, your suggestions
8590 for improvement of @value{GDBN} are welcome in any case.
8594 @section How to report bugs
8596 @cindex @value{GDBN} bugs, reporting
8598 A number of companies and individuals offer support for @sc{gnu} products.
8599 If you obtained @value{GDBN} from a support organization, we recommend you
8600 contact that organization first.
8602 You can find contact information for many support companies and
8603 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
8606 In any event, we also recommend that you send bug reports for @value{GDBN} to one
8610 bug-gdb@@prep.ai.mit.edu
8611 @{ucbvax|mit-eddie|uunet@}!prep.ai.mit.edu!bug-gdb
8614 @strong{Do not send bug reports to @samp{info-gdb}, or to
8615 @samp{help-gdb}, or to any newsgroups.} Most users of @value{GDBN} do not want to
8616 receive bug reports. Those that do have arranged to receive @samp{bug-gdb}.
8618 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
8619 serves as a repeater. The mailing list and the newsgroup carry exactly
8620 the same messages. Often people think of posting bug reports to the
8621 newsgroup instead of mailing them. This appears to work, but it has one
8622 problem which can be crucial: a newsgroup posting often lacks a mail
8623 path back to the sender. Thus, if we need to ask for more information,
8624 we may be unable to reach you. For this reason, it is better to send
8625 bug reports to the mailing list.
8627 As a last resort, send bug reports on paper to:
8630 @sc{gnu} Debugger Bugs
8631 Free Software Foundation Inc.
8632 59 Temple Place - Suite 330
8633 Boston, MA 02111-1307
8637 The fundamental principle of reporting bugs usefully is this:
8638 @strong{report all the facts}. If you are not sure whether to state a
8639 fact or leave it out, state it!
8641 Often people omit facts because they think they know what causes the
8642 problem and assume that some details do not matter. Thus, you might
8643 assume that the name of the variable you use in an example does not matter.
8644 Well, probably it does not, but one cannot be sure. Perhaps the bug is a
8645 stray memory reference which happens to fetch from the location where that
8646 name is stored in memory; perhaps, if the name were different, the contents
8647 of that location would fool the debugger into doing the right thing despite
8648 the bug. Play it safe and give a specific, complete example. That is the
8649 easiest thing for you to do, and the most helpful.
8651 Keep in mind that the purpose of a bug report is to enable us to fix
8652 the bug if it is new to us.
8654 @c FIX ME!!--What the heck does the following sentence mean,
8655 @c in the context of the one above?
8657 @c It is not as important as what happens if the bug is already known.
8659 Therefore, always write your bug reports on
8660 the assumption that the bug has not been reported previously.
8662 Sometimes people give a few sketchy facts and ask, ``Does this ring a
8663 bell?'' Those bug reports are useless, and we urge everyone to
8664 @emph{refuse to respond to them} except to chide the sender to report
8667 To enable us to fix the bug, you should include all these things:
8671 The version of @value{GDBN}. @value{GDBN} announces it if you start with no
8672 arguments; you can also print it at any time using @code{show version}.
8674 Without this, we will not know whether there is any point in looking for
8675 the bug in the current version of @value{GDBN}.
8678 The type of machine you are using, and the operating system name and
8682 What compiler (and its version) was used to compile @value{GDBN}---e.g.
8683 ``@value{GCC}--2.0''.
8686 What compiler (and its version) was used to compile the program you
8687 are debugging---e.g. ``@value{GCC}--2.0''.
8690 The command arguments you gave the compiler to compile your example and
8691 observe the bug. For example, did you use @samp{-O}? To guarantee
8692 you will not omit something important, list them all. A copy of the
8693 Makefile (or the output from make) is sufficient.
8695 If we were to try to guess the arguments, we would probably guess wrong
8696 and then we might not encounter the bug.
8699 A complete input script, and all necessary source files, that will
8703 A description of what behavior you observe that you believe is
8704 incorrect. For example, ``It gets a fatal signal.''
8706 Of course, if the bug is that @value{GDBN} gets a fatal signal, then we will
8707 certainly notice it. But if the bug is incorrect output, we might not
8708 notice unless it is glaringly wrong. You might as well not give us a
8709 chance to make a mistake.
8711 Even if the problem you experience is a fatal signal, you should still
8712 say so explicitly. Suppose something strange is going on, such as,
8713 your copy of @value{GDBN} is out of synch, or you have encountered a
8714 bug in the C library on your system. (This has happened!) Your copy
8715 might crash and ours would not. If you told us to expect a crash,
8716 then when ours fails to crash, we would know that the bug was not
8717 happening for us. If you had not told us to expect a crash, then we
8718 would not be able to draw any conclusion from our observations.
8721 If you wish to suggest changes to the @value{GDBN} source, send us context
8722 diffs. If you even discuss something in the @value{GDBN} source, refer to
8723 it by context, not by line number.
8725 The line numbers in our development sources will not match those in your
8726 sources. Your line numbers would convey no useful information to us.
8729 Here are some things that are not necessary:
8733 A description of the envelope of the bug.
8735 Often people who encounter a bug spend a lot of time investigating
8736 which changes to the input file will make the bug go away and which
8737 changes will not affect it.
8739 This is often time consuming and not very useful, because the way we
8740 will find the bug is by running a single example under the debugger
8741 with breakpoints, not by pure deduction from a series of examples.
8742 We recommend that you save your time for something else.
8744 Of course, if you can find a simpler example to report @emph{instead}
8745 of the original one, that is a convenience for us. Errors in the
8746 output will be easier to spot, running under the debugger will take
8747 less time, and so on.
8749 However, simplification is not vital; if you do not want to do this,
8750 report the bug anyway and send us the entire test case you used.
8753 A patch for the bug.
8755 A patch for the bug does help us if it is a good one. But do not omit
8756 the necessary information, such as the test case, on the assumption that
8757 a patch is all we need. We might see problems with your patch and decide
8758 to fix the problem another way, or we might not understand it at all.
8760 Sometimes with a program as complicated as @value{GDBN} it is very hard to
8761 construct an example that will make the program follow a certain path
8762 through the code. If you do not send us the example, we will not be able
8763 to construct one, so we will not be able to verify that the bug is fixed.
8765 And if we cannot understand what bug you are trying to fix, or why your
8766 patch should be an improvement, we will not install it. A test case will
8767 help us to understand.
8770 A guess about what the bug is or what it depends on.
8772 Such guesses are usually wrong. Even we cannot guess right about such
8773 things without first using the debugger to find the facts.
8776 @c The readline documentation is distributed with the readline code
8777 @c and consists of the two following files:
8780 @c Use -I with makeinfo to point to the appropriate directory,
8781 @c environment var TEXINPUTS with TeX.
8782 @include rluser.texinfo
8783 @include inc-hist.texi
8787 @node Renamed Commands
8788 @appendix Renamed Commands
8790 The following commands were renamed in @value{GDBN} 4, in order to make the
8791 command set as a whole more consistent and easier to use and remember:
8794 @kindex delete environment
8795 @kindex info copying
8796 @kindex info convenience
8797 @kindex info directories
8798 @kindex info editing
8799 @kindex info history
8800 @kindex info targets
8802 @kindex info version
8803 @kindex info warranty
8804 @kindex set addressprint
8805 @kindex set arrayprint
8806 @kindex set prettyprint
8807 @kindex set screen-height
8808 @kindex set screen-width
8809 @kindex set unionprint
8810 @kindex set vtblprint
8811 @kindex set demangle
8812 @kindex set asm-demangle
8813 @kindex set sevenbit-strings
8814 @kindex set array-max
8816 @kindex set history write
8817 @kindex show addressprint
8818 @kindex show arrayprint
8819 @kindex show prettyprint
8820 @kindex show screen-height
8821 @kindex show screen-width
8822 @kindex show unionprint
8823 @kindex show vtblprint
8824 @kindex show demangle
8825 @kindex show asm-demangle
8826 @kindex show sevenbit-strings
8827 @kindex show array-max
8828 @kindex show caution
8829 @kindex show history write
8834 @c END TEXI2ROFF-KILL
8836 OLD COMMAND NEW COMMAND
8838 --------------- -------------------------------
8839 @c END TEXI2ROFF-KILL
8840 add-syms add-symbol-file
8841 delete environment unset environment
8842 info convenience show convenience
8843 info copying show copying
8844 info directories show directories
8845 info editing show commands
8846 info history show values
8847 info targets help target
8848 info values show values
8849 info version show version
8850 info warranty show warranty
8851 set/show addressprint set/show print address
8852 set/show array-max set/show print elements
8853 set/show arrayprint set/show print array
8854 set/show asm-demangle set/show print asm-demangle
8855 set/show caution set/show confirm
8856 set/show demangle set/show print demangle
8857 set/show history write set/show history save
8858 set/show prettyprint set/show print pretty
8859 set/show screen-height set/show height
8860 set/show screen-width set/show width
8861 set/show sevenbit-strings set/show print sevenbit-strings
8862 set/show unionprint set/show print union
8863 set/show vtblprint set/show print vtbl
8865 unset [No longer an alias for delete]
8871 \vskip \parskip\vskip \baselineskip
8872 \halign{\tt #\hfil &\qquad#&\tt #\hfil\cr
8873 {\bf Old Command} &&{\bf New Command}\cr
8874 add-syms &&add-symbol-file\cr
8875 delete environment &&unset environment\cr
8876 info convenience &&show convenience\cr
8877 info copying &&show copying\cr
8878 info directories &&show directories \cr
8879 info editing &&show commands\cr
8880 info history &&show values\cr
8881 info targets &&help target\cr
8882 info values &&show values\cr
8883 info version &&show version\cr
8884 info warranty &&show warranty\cr
8885 set{\rm / }show addressprint &&set{\rm / }show print address\cr
8886 set{\rm / }show array-max &&set{\rm / }show print elements\cr
8887 set{\rm / }show arrayprint &&set{\rm / }show print array\cr
8888 set{\rm / }show asm-demangle &&set{\rm / }show print asm-demangle\cr
8889 set{\rm / }show caution &&set{\rm / }show confirm\cr
8890 set{\rm / }show demangle &&set{\rm / }show print demangle\cr
8891 set{\rm / }show history write &&set{\rm / }show history save\cr
8892 set{\rm / }show prettyprint &&set{\rm / }show print pretty\cr
8893 set{\rm / }show screen-height &&set{\rm / }show height\cr
8894 set{\rm / }show screen-width &&set{\rm / }show width\cr
8895 set{\rm / }show sevenbit-strings &&set{\rm / }show print sevenbit-strings\cr
8896 set{\rm / }show unionprint &&set{\rm / }show print union\cr
8897 set{\rm / }show vtblprint &&set{\rm / }show print vtbl\cr
8899 unset &&\rm(No longer an alias for delete)\cr
8902 @c END TEXI2ROFF-KILL
8906 @ifclear PRECONFIGURED
8907 @node Formatting Documentation
8908 @appendix Formatting Documentation
8910 @cindex @value{GDBN} reference card
8911 @cindex reference card
8912 The @value{GDBN} 4 release includes an already-formatted reference card, ready
8913 for printing with PostScript or Ghostscript, in the @file{gdb}
8914 subdirectory of the main source directory@footnote{In
8915 @file{gdb-@value{GDBVN}/gdb/refcard.ps} of the version @value{GDBVN}
8916 release.}. If you can use PostScript or Ghostscript with your printer,
8917 you can print the reference card immediately with @file{refcard.ps}.
8919 The release also includes the source for the reference card. You
8920 can format it, using @TeX{}, by typing:
8926 The @value{GDBN} reference card is designed to print in @dfn{landscape}
8927 mode on US ``letter'' size paper;
8928 that is, on a sheet 11 inches wide by 8.5 inches
8929 high. You will need to specify this form of printing as an option to
8930 your @sc{dvi} output program.
8932 @cindex documentation
8934 All the documentation for @value{GDBN} comes as part of the machine-readable
8935 distribution. The documentation is written in Texinfo format, which is
8936 a documentation system that uses a single source file to produce both
8937 on-line information and a printed manual. You can use one of the Info
8938 formatting commands to create the on-line version of the documentation
8939 and @TeX{} (or @code{texi2roff}) to typeset the printed version.
8941 @value{GDBN} includes an already formatted copy of the on-line Info version of
8942 this manual in the @file{gdb} subdirectory. The main Info file is
8943 @file{gdb-@r{version-number}/gdb/gdb.info}, and it refers to
8944 subordinate files matching @samp{gdb.info*} in the same directory. If
8945 necessary, you can print out these files, or read them with any editor;
8946 but they are easier to read using the @code{info} subsystem in @sc{gnu} Emacs
8947 or the standalone @code{info} program, available as part of the @sc{gnu}
8948 Texinfo distribution.
8950 If you want to format these Info files yourself, you need one of the
8951 Info formatting programs, such as @code{texinfo-format-buffer} or
8954 If you have @code{makeinfo} installed, and are in the top level @value{GDBN}
8955 source directory (@file{gdb-@value{GDBVN}}, in the case of version @value{GDBVN}), you can
8956 make the Info file by typing:
8963 If you want to typeset and print copies of this manual, you need @TeX{},
8964 a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the
8965 Texinfo definitions file.
8967 @TeX{} is a typesetting program; it does not print files directly, but
8968 produces output files called @sc{dvi} files. To print a typeset
8969 document, you need a program to print @sc{dvi} files. If your system
8970 has @TeX{} installed, chances are it has such a program. The precise
8971 command to use depends on your system; @kbd{lpr -d} is common; another
8972 (for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may
8973 require a file name without any extension or a @samp{.dvi} extension.
8975 @TeX{} also requires a macro definitions file called
8976 @file{texinfo.tex}. This file tells @TeX{} how to typeset a document
8977 written in Texinfo format. On its own, @TeX{} cannot either read or
8978 typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB
8979 and is located in the @file{gdb-@var{version-number}/texinfo}
8982 If you have @TeX{} and a @sc{dvi} printer program installed, you can
8983 typeset and print this manual. First switch to the the @file{gdb}
8984 subdirectory of the main source directory (for example, to
8985 @file{gdb-@value{GDBVN}/gdb}) and then type:
8991 @node Installing GDB
8992 @appendix Installing @value{GDBN}
8993 @cindex configuring @value{GDBN}
8994 @cindex installation
8996 @value{GDBN} comes with a @code{configure} script that automates the process
8997 of preparing @value{GDBN} for installation; you can then use @code{make} to
8998 build the @code{gdb} program.
9000 @c irrelevant in info file; it's as current as the code it lives with.
9001 @footnote{If you have a more recent version of @value{GDBN} than @value{GDBVN},
9002 look at the @file{README} file in the sources; we may have improved the
9003 installation procedures since publishing this manual.}
9006 The @value{GDBN} distribution includes all the source code you need for
9007 @value{GDBN} in a single directory, whose name is usually composed by
9008 appending the version number to @samp{gdb}.
9010 For example, the @value{GDBN} version @value{GDBVN} distribution is in the
9011 @file{gdb-@value{GDBVN}} directory. That directory contains:
9014 @item gdb-@value{GDBVN}/configure @r{(and supporting files)}
9015 script for configuring @value{GDBN} and all its supporting libraries
9017 @item gdb-@value{GDBVN}/gdb
9018 the source specific to @value{GDBN} itself
9020 @item gdb-@value{GDBVN}/bfd
9021 source for the Binary File Descriptor library
9023 @item gdb-@value{GDBVN}/include
9024 @sc{gnu} include files
9026 @item gdb-@value{GDBVN}/libiberty
9027 source for the @samp{-liberty} free software library
9029 @item gdb-@value{GDBVN}/opcodes
9030 source for the library of opcode tables and disassemblers
9032 @item gdb-@value{GDBVN}/readline
9033 source for the @sc{gnu} command-line interface
9035 @item gdb-@value{GDBVN}/glob
9036 source for the @sc{gnu} filename pattern-matching subroutine
9038 @item gdb-@value{GDBVN}/mmalloc
9039 source for the @sc{gnu} memory-mapped malloc package
9042 The simplest way to configure and build @value{GDBN} is to run @code{configure}
9043 from the @file{gdb-@var{version-number}} source directory, which in
9044 this example is the @file{gdb-@value{GDBVN}} directory.
9046 First switch to the @file{gdb-@var{version-number}} source directory
9047 if you are not already in it; then run @code{configure}. Pass the
9048 identifier for the platform on which @value{GDBN} will run as an
9054 cd gdb-@value{GDBVN}
9055 ./configure @var{host}
9060 where @var{host} is an identifier such as @samp{sun4} or
9061 @samp{decstation}, that identifies the platform where @value{GDBN} will run.
9062 (You can often leave off @var{host}; @code{configure} tries to guess the
9063 correct value by examining your system.)
9065 Running @samp{configure @var{host}} and then running @code{make} builds the
9066 @file{bfd}, @file{readline}, @file{mmalloc}, and @file{libiberty}
9067 libraries, then @code{gdb} itself. The configured source files, and the
9068 binaries, are left in the corresponding source directories.
9071 @code{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
9072 system does not recognize this automatically when you run a different
9073 shell, you may need to run @code{sh} on it explicitly:
9076 sh configure @var{host}
9079 If you run @code{configure} from a directory that contains source
9080 directories for multiple libraries or programs, such as the
9081 @file{gdb-@value{GDBVN}} source directory for version @value{GDBVN}, @code{configure}
9082 creates configuration files for every directory level underneath (unless
9083 you tell it not to, with the @samp{--norecursion} option).
9085 You can run the @code{configure} script from any of the
9086 subordinate directories in the @value{GDBN} distribution if you only want to
9087 configure that subdirectory, but be sure to specify a path to it.
9089 For example, with version @value{GDBVN}, type the following to configure only
9090 the @code{bfd} subdirectory:
9094 cd gdb-@value{GDBVN}/bfd
9095 ../configure @var{host}
9099 You can install @code{@value{GDBP}} anywhere; it has no hardwired paths.
9100 However, you should make sure that the shell on your path (named by
9101 the @samp{SHELL} environment variable) is publicly readable. Remember
9102 that @value{GDBN} uses the shell to start your program---some systems refuse to
9103 let @value{GDBN} debug child processes whose programs are not readable.
9106 * Separate Objdir:: Compiling @value{GDBN} in another directory
9107 * Config Names:: Specifying names for hosts and targets
9108 * configure Options:: Summary of options for configure
9111 @node Separate Objdir
9112 @section Compiling @value{GDBN} in another directory
9114 If you want to run @value{GDBN} versions for several host or target machines,
9115 you need a different @code{gdb} compiled for each combination of
9116 host and target. @code{configure} is designed to make this easy by
9117 allowing you to generate each configuration in a separate subdirectory,
9118 rather than in the source directory. If your @code{make} program
9119 handles the @samp{VPATH} feature (@sc{gnu} @code{make} does), running
9120 @code{make} in each of these directories builds the @code{gdb}
9121 program specified there.
9123 To build @code{gdb} in a separate directory, run @code{configure}
9124 with the @samp{--srcdir} option to specify where to find the source.
9125 (You also need to specify a path to find @code{configure}
9126 itself from your working directory. If the path to @code{configure}
9127 would be the same as the argument to @samp{--srcdir}, you can leave out
9128 the @samp{--srcdir} option; it is assumed.)
9130 For example, with version @value{GDBVN}, you can build @value{GDBN} in a
9131 separate directory for a Sun 4 like this:
9135 cd gdb-@value{GDBVN}
9138 ../gdb-@value{GDBVN}/configure sun4
9143 When @code{configure} builds a configuration using a remote source
9144 directory, it creates a tree for the binaries with the same structure
9145 (and using the same names) as the tree under the source directory. In
9146 the example, you'd find the Sun 4 library @file{libiberty.a} in the
9147 directory @file{gdb-sun4/libiberty}, and @value{GDBN} itself in
9148 @file{gdb-sun4/gdb}.
9150 One popular reason to build several @value{GDBN} configurations in separate
9151 directories is to configure @value{GDBN} for cross-compiling (where
9152 @value{GDBN} runs on one machine---the @dfn{host}---while debugging
9153 programs that run on another machine---the @dfn{target}).
9154 You specify a cross-debugging target by
9155 giving the @samp{--target=@var{target}} option to @code{configure}.
9157 When you run @code{make} to build a program or library, you must run
9158 it in a configured directory---whatever directory you were in when you
9159 called @code{configure} (or one of its subdirectories).
9161 The @code{Makefile} that @code{configure} generates in each source
9162 directory also runs recursively. If you type @code{make} in a source
9163 directory such as @file{gdb-@value{GDBVN}} (or in a separate configured
9164 directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{GDBVN}}), you
9165 will build all the required libraries, and then build GDB.
9167 When you have multiple hosts or targets configured in separate
9168 directories, you can run @code{make} on them in parallel (for example,
9169 if they are NFS-mounted on each of the hosts); they will not interfere
9173 @section Specifying names for hosts and targets
9175 The specifications used for hosts and targets in the @code{configure}
9176 script are based on a three-part naming scheme, but some short predefined
9177 aliases are also supported. The full naming scheme encodes three pieces
9178 of information in the following pattern:
9181 @var{architecture}-@var{vendor}-@var{os}
9184 For example, you can use the alias @code{sun4} as a @var{host} argument,
9185 or as the value for @var{target} in a @code{--target=@var{target}}
9186 option. The equivalent full name is @samp{sparc-sun-sunos4}.
9188 The @code{configure} script accompanying @value{GDBN} does not provide
9189 any query facility to list all supported host and target names or
9190 aliases. @code{configure} calls the Bourne shell script
9191 @code{config.sub} to map abbreviations to full names; you can read the
9192 script, if you wish, or you can use it to test your guesses on
9193 abbreviations---for example:
9196 % sh config.sub sun4
9197 sparc-sun-sunos4.1.1
9198 % sh config.sub sun3
9200 % sh config.sub decstation
9202 % sh config.sub hp300bsd
9204 % sh config.sub i386v
9206 % sh config.sub i786v
9207 Invalid configuration `i786v': machine `i786v' not recognized
9211 @code{config.sub} is also distributed in the @value{GDBN} source
9212 directory (@file{gdb-@value{GDBVN}}, for version @value{GDBVN}).
9214 @node configure Options
9215 @section @code{configure} options
9217 Here is a summary of the @code{configure} options and arguments that
9218 are most often useful for building @value{GDBN}. @code{configure} also has
9219 several other options not listed here. @inforef{What Configure
9220 Does,,configure.info}, for a full explanation of @code{configure}.
9223 configure @r{[}--help@r{]}
9224 @r{[}--prefix=@var{dir}@r{]}
9225 @r{[}--srcdir=@var{dirname}@r{]}
9226 @r{[}--norecursion@r{]} @r{[}--rm@r{]}
9227 @r{[}--target=@var{target}@r{]} @var{host}
9231 You may introduce options with a single @samp{-} rather than
9232 @samp{--} if you prefer; but you may abbreviate option names if you use
9237 Display a quick summary of how to invoke @code{configure}.
9239 @item -prefix=@var{dir}
9240 Configure the source to install programs and files under directory
9243 @c avoid splitting the warning from the explanation:
9245 @item --srcdir=@var{dirname}
9246 @strong{Warning: using this option requires @sc{gnu} @code{make}, or another
9247 @code{make} that implements the @code{VPATH} feature.}@*
9248 Use this option to make configurations in directories separate from the
9249 @value{GDBN} source directories. Among other things, you can use this to
9250 build (or maintain) several configurations simultaneously, in separate
9251 directories. @code{configure} writes configuration specific files in
9252 the current directory, but arranges for them to use the source in the
9253 directory @var{dirname}. @code{configure} creates directories under
9254 the working directory in parallel to the source directories below
9258 Configure only the directory level where @code{configure} is executed; do not
9259 propagate configuration to subdirectories.
9262 @emph{Remove} files otherwise built during configuration.
9264 @c This does not work (yet if ever). FIXME.
9265 @c @item --parse=@var{lang} @dots{}
9266 @c Configure the @value{GDBN} expression parser to parse the listed languages.
9267 @c @samp{all} configures @value{GDBN} for all supported languages. To get a
9268 @c list of all supported languages, omit the argument. Without this
9269 @c option, @value{GDBN} is configured to parse all supported languages.
9271 @item --target=@var{target}
9272 Configure @value{GDBN} for cross-debugging programs running on the specified
9273 @var{target}. Without this option, @value{GDBN} is configured to debug
9274 programs that run on the same machine (@var{host}) as @value{GDBN} itself.
9276 There is no convenient way to generate a list of all available targets.
9278 @item @var{host} @dots{}
9279 Configure @value{GDBN} to run on the specified @var{host}.
9281 There is no convenient way to generate a list of all available hosts.
9285 @code{configure} accepts other options, for compatibility with
9286 configuring other @sc{gnu} tools recursively; but these are the only
9287 options that affect @value{GDBN} or its supporting libraries.
9296 % I think something like @colophon should be in texinfo. In the
9298 \long\def\colophon{\hbox to0pt{}\vfill
9299 \centerline{The body of this manual is set in}
9300 \centerline{\fontname\tenrm,}
9301 \centerline{with headings in {\bf\fontname\tenbf}}
9302 \centerline{and examples in {\tt\fontname\tentt}.}
9303 \centerline{{\it\fontname\tenit\/},}
9304 \centerline{{\bf\fontname\tenbf}, and}
9305 \centerline{{\sl\fontname\tensl\/}}
9306 \centerline{are used for emphasis.}\vfill}
9308 % Blame: doc@cygnus.com, 1991.